Patent Grant
9081879
1. Field of the Invention
The field of the invention relates to computerized medical diagnostic systems. More particularly, embodiments of the present invention relate to a computerized system and method for diagnosing a patient's medical complaint by utilizing the affected anatomic system and probable cause of the patient's medical complaint.
2. Description of the Related Art
Emergency physicians often have very little time and often even less information about a patient who is brought to the emergency department. Many errors occur because the physician may not consider all of the possible causes of the patient's complaints. A list of the possible diagnoses that a patient can have is called a differential diagnosis.
A few examples will help define the problem. For instance, a patient may be brought in comatose with a high fever. The emergency physician needs to know immediately, sometimes within seconds the causes of coma and particularly the causes of coma that could be caused by infection.
For example, a patient may exhibit a cough and a fever. A logical assumption would be that the patient has an infection (cause) in the respiratory system (anatomic system).
For example, a patient may be brought to the ER in a coma and there may be no history other than that he works in a garage indicating possible carbon monoxide poisoning.
Further, the best testing strategy to distinguish between several possible diagnoses has to be planned in, again, very little time.
What is desired is a way to allow a clinician to start an evaluation based on the limited information that is available. This frequently occurs in emergency medicine, where there are significant constraints on time and information.
There are other occasions when medical personnel have more time to obtain detailed information from and about the patient, and/or the patient can provide additional information by themselves. In these circumstances, a detailed fully automated approach to developing a diagnosis is desired. There are yet other occasions when various combinations of the above are desired depending on the time and information available.
In one embodiment of the invention, there is an automated medical diagnostic system, comprising a graphical user interface (GUI) comprising a matrix of cells, each cell being representative of a patient health item (PHI) associated with a particular disease; a plurality of diagnostic functions that uses the matrix for input and output, wherein the diagnostic functions are selectively executed according to each newly input PHI so as to result in and display a differential diagnosis.
A disease may be an abnormality, affliction, ailment, anomaly, disorder, illness, indisposition, infirmity, malady, problem or sickness of a patent. The system may use a Meta analysis of the input PHIs.
The system may use a reenter function so as to request a user of the system to consult the system at a later time.
A PHI may be a sign, symptom, complaint, presentation, manifestation, finding, laboratory test result, home test result, or an interpretation of an imaging study.
The GUI may be configured for use by a trained health professional.
The system may be configured for use by physicians, physician's assistants, nurses, paramedics, emergency medical technicians, veterinarians, chiropractors, dentists, optometrists, and holistic health providers.
The system may use software objects.
Scoring and thresholding may be used by at least one of the diagnostic functions.
The differential diagnosis may be a single disease.
The diagnostic functions may form a list-based processing system.
The diagnostic functions may use a patient medical record.
The system may run on a handheld computing device, a personal digital assistant (PDA) or similar computing device.
The onset time shape of a PHI may be captured in a graphical manner in which the x axis represents time and the y axis may represent severity.
The change in intensity of a PHI over time may be graphically displayed on the GUI.
The variability of any aspect of a PHI can be expressed graphically and this image transformed into information to diagnose a disease.
The interrelationships of PHIs over time may be input from a Gantt chart based input system.
The value of one PHI may be varied by a process of sensitivity analysis, producing changes in the differential diagnosis of a set of PHIs.
The total amount of time the patient has presented a PHI or condition may be input graphically and the area under a corresponding curve may be determined by integrating the graphed function.
When a patient indicates their condition begins to improve, the system may cause a specific action or may stop a particular consultation to allow more time to elapse to determine if a trend continues.
The topographic area of a patient's complaint(s) may be indicated on an avatar displayed by the GUI which then causes different diagnostic weights to be added to various diseases.
The sensitivity of a consultation can be varied by levels such that a number and area of specialty interest of diseases considered in a consultation can be increased in steps from urgent and common diseases to all diseases known to mankind. The matrix may be used to specify the levels and the area of specialty interest.
All human diseases stored in the system may be considered in the differential diagnosis.
A patient's incidental symptoms may be saved and evaluated at a later time in a different context.
An ongoing disease process may be distinguished from a disease process in which the symptoms have disappeared.
Each PHI may have associated questions to establish the PHI, and diagnosis may occur in two different modes, including a horizontal axis of inquiry (HAI) mode in which the questions for each PHI may be selected with the understanding that several diseases are under diagnostic consideration, and a vertical axis of inquiry (VAI) mode in which one disease may be the source of all the questions for each PHI or in which one disease may be the focus of inquiry. Diseases in the HAI may be segregated into a subset for further focused evaluation in which all questions may be taken from the subset of diseases. The system may start operation in HAI mode and then turn to VAI mode upon certain criteria being met, or it may start in VAI mode, usually with the most urgent diagnosis under consideration.
The diseases may be weighted if a PHI matches.
The attributes of a PHI may be evaluated to further define the PHI and in which diagnostic weight may be sequentially added depending upon the number of matching attributes.
Each PHI in the system may be represented by a portion of an inverted tree with a basic and general PHI at the root of the tree and a detailed and specific PHI or PHI attribute at the leaves of the tree. Diagnostic weight may be sequentially added as a more precise match is made as the questioning proceeds down the tree. PHIs may be hierarchically weighted based upon their position in the tree.
The relationships between risk factors, causes, and complications of a disease may be graphically displayed, with the x-axis being time, by the GUI.
The interrelated causality of diseases may be expressed graphically and dynamically.
Geographical information related to the PHIs may be entered on a graphic image of a world map.
A patient work up may be automatically planned to diagnose one disease from a set of possible diseases.
Questions may be presented directly to the patient. An interactive Gantt chart may be used to express the relationships between various PHIs in a disease with respect to time and that information may be used to diagnose the patient.
A disease may be defined or represented by a Gantt chart displayed by the GUI.
One or more predetermined PHIs may be displayed and immediately selected if they should occur during an interview process of a consultation which may cause a change in the execution path of the system or may cause another action to take place.
Potentially dangerous situations may be automatically recognized and a metric of the potential danger to the patient may be graphically displayed by the GUI.
Potentially dangerous situations may be automatically recognized and a metric of the potential danger to the care giver may be graphically displayed by the GUI.
A feedback loop may exist in which at least one diagnostic weight corresponding to at least one PHI may be automatically changed based on data obtained by usage of the system over time.
In another embodiment of the invention, there is an automated medical diagnostic method, comprising displaying a matrix of cells, each cell being representative of a patient health item (PHI) associated with a particular disease; entering a PHI on the display in one of the cells; updating a differential diagnosis based on the input PHI; and displaying the updated differential diagnosis.
The displaying may be performed via a graphical user interface (GUI). The GUI may be configured for use by a trained health professional.
A PHI may be a sign, symptom, complaint, presentation, manifestation, finding, laboratory test result, home test result, or an interpretation of an imaging study.
The method may be configured for use by veterinarians, chiropractors, dentists, optometrists, or holistic health providers.
The updated differential diagnosis may be a single disease.
The method may additionally comprise storing the input PHI in a patient medical record.
The method may be performed on a handheld computing device, a personal digital assistant (PDA) or a PDA-like computing device.
The method may additionally comprise presenting questions corresponding to at least one PHI directly to the patient.
In another embodiment of the matrix, the X-axis presents standard anatomic systems and the Y-axis shows the different parts of a workup. These being, chief complaints, history of the present illness, past medical history, family medical history, social history, vital signs, physical examination findings, lab tests, special studies, imaging, and treatment. By selecting a cell, for example, the cell at the intersection of respiratory system and chief complaints, all of the chief complaints related to the respiratory system would be displayed.
The following description presents certain specific embodiments of the present invention. However, the present invention may be embodied in a multitude of different ways as defined and covered by the claims. In this description, reference is made to the drawings wherein like parts are designated with like numerals throughout.
A matrix-medical diagnostic and treatment advice (MDATA) embodiment of the First Opinion fully automated patient specific diagnostic and treatment advice system will be described. The matrix-MDATA (MM) system runs on any computer but is designed to be used by a healthcare giver during a bedside interaction with a patient. An exemplary configuration of components is described hereinbelow.
The MM system is designed to provide the correct diagnosis at the earliest possible point in time while asking the fewest number of questions. The MM system combines the MDATA system, such as described in U.S. Pat. Nos. 5,660,176, 5,935,060, 6,022,315, and 6,527,713, with the Matrix system, such as described in U.S. patent application Ser. No. 10/862,116, and which are all hereby incorporated by reference.
The MM system may be used in a fully automated mode, reflecting the MDATA system, or in a health care giver mode, reflecting the Matrix system, or any combination of the two modes.
The MM system is comprised of various modules, tools, and applications as discussed in detail below. As can be appreciated by one of ordinary skill in the art, each of the modules may comprise various sub-routines, procedures, definitional statements and macros. Each of the modules are typically separately compiled and linked into a single executable program. Therefore, the following description of each of the modules is used for convenience to describe the functionality of the preferred system. Thus, the processes that are undergone by each of the modules may be arbitrarily redistributed to one of the other modules, combined together in a single module, or made available in, for example, a shareable dynamic link library.
The system modules, tools, and applications may be written in any programming language such as, for example, C, C++, BASIC, Visual Basic, Pascal, Ada, Java, HTML, XML, or FORTRAN, and executed on an operating system, such as variants of Windows, Macintosh, UNIX, Linux, VxWorks, or other operating system. C, C++, BASIC, Visual Basic, Pascal, Ada, Java, HTML, XML and FORTRAN are industry standard programming languages for which many commercial compilers can be used to create executable code.
Definitions
The following provides a number of useful possible definitions of terms used in describing certain embodiments of the disclosed invention.
A network may refer to a network or combination of networks spanning any geographical area, such as a local area network (LAN), wide area network (WAN), regional network, national network, and/or global network. The Internet is an example of a current global computer network. Those terms may refer to hardwire networks, wireless networks, or a combination of hardwire and wireless networks. Hardwire networks may include, for example, fiber optic lines, cable lines, ISDN lines, copper lines, etc. Wireless networks may include, for example, cellular systems, personal communications service (PCS) systems, satellite communication systems, packet radio systems, and mobile broadband systems. A cellular system may use, for example, code division multiple access (CDMA), time division multiple access (TDMA), personal digital phone (PDC), Global System Mobile (GSM), or frequency division multiple access (FDMA), among others.
A website may refer to one or more interrelated web page files and other files and programs on one or more web servers. The files and programs are accessible over a computer network, such as the Internet, by sending a hypertext transfer protocol (HTTP or HTTPS [S-HTTP]) request specifying a uniform resource locator (URL) that identifies the location of one of said web page files, wherein the files and programs are owned, managed or authorized by a single business entity. Such files and programs can include, for example, hypertext markup language (HTML) files, common gateway interface (CGI) files, and Java applications. The web page files preferably include a home page file that corresponds to a home page of the website. The home page can serve as a gateway or access point to the remaining files and programs contained within the website. In one embodiment, all of the files and programs are located under, and accessible within, the same network domain as the home page file. Alternatively, the files and programs can be located and accessible through several different network domains.
A web page or electronic page may comprise that which is presented by a standard web browser in response to an HTTP request specifying the URL by which the web page file is identified. A web page can include, for example, text, images, sound, video, and animation.
A computer or computing device may be any processor controlled device that permits access to the Internet, including terminal devices, such as personal computers, workstations, servers, clients, mini-computers, main-frame computers, laptop computers, a network of individual computers, mobile computers, palm-top computers, hand-held computers, set top boxes for a television, other types of web-enabled televisions, interactive kiosks, personal digital assistants (PDAs), interactive or web-enabled wireless communications devices, mobile web browsers, or a combination thereof. The computers may further possess one or more input devices such as a keyboard, mouse, touch pad, joystick, pen-input-pad, and the like. The computers may also possess an output device, such as a visual display and an audio output. One or more of these computing devices may form a computing environment.
These computers may be uni-processor or multi-processor machines. Additionally, these computers may include an addressable storage medium or computer accessible medium, such as random access memory (RAM), an electronically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), hard disks, floppy disks, laser disk players, digital video devices, compact disks, video tapes, audio tapes, magnetic recording tracks, electronic networks, and other techniques to transmit or store electronic content such as, by way of example, programs and data. In one embodiment, the computers are equipped with a network communication device such as a network interface card, a modem, or other network connection device suitable for connecting to the communication network. Furthermore, the computers execute an appropriate operating system such as Linux, UNIX, any of the versions of Microsoft Windows, Apple MacOS, IBM OS/2 or other operating system. The appropriate operating system may include a communications protocol implementation that handles all incoming and outgoing message traffic passed over the network. In other embodiments, while the operating system may differ depending on the type of computer, the operating system will continue to provide the appropriate communications protocols to establish communication links with the network.
The computers may contain program logic, or other substrate configuration representing data and instructions, which cause the computer to operate in a specific and predefined manner, as described herein. In one embodiment, the program logic may be implemented as one or more object frameworks or modules. These modules may be configured to reside on the addressable storage medium and configured to execute on one or more processors. The modules include, but are not limited to, software or hardware components that perform certain tasks. Thus, a module may include, by way of example, components, such as, software components, object-oriented software components, class components and task components, processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables.
The various components of the system may communicate with each other and other components comprising the respective computers through mechanisms such as, by way of example, interprocess communication, remote procedure call, distributed object interfaces, and other various program interfaces. Furthermore, the functionality provided for in the components, modules, and databases may be combined into fewer components, modules, or databases or further separated into additional components, modules, or databases. Additionally, the components, modules, and databases may be implemented to execute on one or more computers.
Exemplary MM System Configuration
Referring to
When the computing device 2610 is connected with the server(s) 2650, the web site may optionally provide updates on new disease information or about new laboratory tests, special studies, imaging modality of choice and treatment of choice. In addition, the computing device 2610 can optionally be linked to the network 2640 to allow instantaneous reporting of and downloading information about, for example, possible epidemics or the use of weapons of mass destruction (WMD). In another embodiment, the computing device 2610 runs only when connected to the server(s) 2650.
The computing device 2610 includes a processor 2612, a display 2614, and one or more input devices 2616. The processor 2612 is in data communication with a data storage 2618 for storing one or more databases having medical data used by the system. In certain embodiments, the data storage 2618 stores patient medical records. System software 2620 is executed by the processor 2612. The system software 2620 includes an application graphical user interface (GUI). The application GUI can include a database interface to the data storage 2618 of the computing device. In certain embodiments, the software is loaded from the data storage 2618. In embodiments where the computing device 2610 communicates with a web site, the processor utilizes browser software in place of or in addition to the software 2620.
Overview of Object-Based Diagnosis
In computer software terms, an object is combination of data and processes that manipulate the data. The data are said to be “encapsulated,” meaning that the data is hidden, so that a user of the object only sees processes that can be invoked. Using an object's processes, one can then manipulate the data without having to know the exact location and format of the data. When more than one copy of the object is required, one can make copies of the data, but use the same process set to manipulate each of the copies as needed. This set of processes can then be thought of as an “engine” that controls or represents the objects' behavior, whether there are 10 or 10,000 object copies.
This section describes a diagnostic paradigm that uses software objects to establish a broad, generalized software environment for medical diagnosis, which is used to define and develop the programming elements of medical diagnosis. The objects are then used to guide and control the diagnostic process, to conduct the patient interview, to perform related analytical tasks, and to generate the diagnoses. A software object is a fundamental software structure that can be used to organize the processes and data of a computer program in such a way as to make possible very complicated applications. This description will discuss novel uses of object oriented programming (OOP) in medical diagnosis, such as the use of software objects for the purpose of fully automated medical diagnosis, the entire/overall method of dynamically assembling the components of diagnosis in the form of objects, and then letting the objects interact to compute a diagnosis.
Defining and creating software objects is well-known to any programmer trained in object-oriented programming. Using an OOP-capable compiler, the programmer defines the data that represent the object and the actions that the object can perform. At run time, the program creates an object, supplies the data that define the object, and then manipulates the object using the object actions. The program can create any number of objects as needed. Each object can be independently initialized, manipulated, and destroyed.
In the object-based (OB) method discussed here, software objects are used as active, intelligent agents that represent all of the players and all of the data in suitably organized roles. It is important to note in this metaphor that all of the disease objects, which are “specialists” for a single disease, are allowed to monitor the questions and answers of other objects.
One key concept of the OB method is to think of disease and symptom objects as representing the medical experts inside the computer. If we ask the Appendicitis Disease Object to look at a patient, the object looks at the patient data, notes that the patient does indeed complain of abdominal pain and nausea—but then “notices” the appendectomy scar! This indicates that appendicitis can probably be ruled out, so VAI would step down and return control to HAI so that another more likely diagnosis can be considered first. In this case HAI might invoke another VAI expert, say, Small Bowel Obstruction. That object takes a look, asks some questions, and renders its opinion on the probability of it being the diagnosis. In effect, a huge number of diagnostic experts are gathered at the patient's bedside, and each object may get a turn at evaluating the patient data in terms of its own symptom pattern.
As an actual patient symptom set is built up, disease objects judge themselves and return a probability that they are the correct diagnosis. The emergent effect is a patient interview and a diagnostic evaluation that—by design—constantly stays focused on the most likely set of diseases of the patient. Carefully focused questions are used to eliminate or reduce the likelihood of diseases, to promote others into the “realm of suspicion,” and to expand the search in a promising direction, based on the data being obtained from the patient.
In this example, the disease object can make time-based recommendations to the system or the patient. This “advice” can then be used as part of a report to the user or the patient as “discharge instructions”.
A software “object” is basically a data structure plus associated processes that can do things with, for or to the data. An important property of an object is that the object's data can be hidden behind the object's processes, so that the outside user of the object can only see and use object processes that can be invoked to access the data. The object is said to “hide” data; it provides the powerful ability of decoupling the world that uses an object from the object itself.
A. Disease Object
A Disease Object (DO) is a software object that represents an abnormal health state (illness, disease, disorder, or cause) which is collectively called a “disease.” It is used in the system and method to establish the likelihood that the specified disease exists in the current patient.
B. Symptom Object
A Symptom Object (SO) is a software object that represents a patient health item (sign, symptom, complaint, presentation, manifestation, finding, laboratory test result (home or remote), and interpretations of an imaging study) which is collectively called a symptom or a PHI (Patient Health Item). It is used in the system to describe patient health in terms that the list-based (LB) system can use for diagnosis. It is important to note that PHI may be a synergy. Synergies are combinations of other PHIs either occurring together or sequentially which make one disease more likely than another. A simultaneous synergy is the occurrence of two or more PHIs occurring at the same time. A sequential synergy occurs when two or more PHIs occur in a certain time sequence. A summation synergy is a combination of other synergies.
C. Valuator Object
A Valuator Object (VO) is a software object that represents the actions required to establish the value of a symptom in a patient at a specified time.
D. Question Object
A Question Object (QO) is a software object that describes the software elements required to establish a mini-dialog of questions and responses with the patient, in order to obtain a symptom value. It is the task of the QO to select the appropriate question set, to invoke the appropriate node objects that actually question the patient, and to report back the patient's response. A QO is a type of valuator object that specializes in interaction with a patient.
E. Node Object
A Node Object (NO) is a software object that describes the software elements required to ask a single, well-defined question of the patient and to return the response selected by the patient. It is the task of the NO to present the required data to the GUI in a form that will appear in a user-friendly manner on the user display, to wait an appropriate amount of time for a user response, to possibly re-prompt the user, and to ultimately return the user's response.
Brief Review of MDATA System
A fear of many patients is that they may have a disease that their doctor is not familiar with. The advantage of an automated system is that it is as easy to consider 2000 diseases as it is to consider 20 diseases. The MM system can be customized to consider only the most urgent, serious and common diseases, or to consider every disease of mankind. While not technically true, the point is clear. It is simply not possible for any human being to be familiar with every disease, let alone be able to diagnose it. But a fully automated system such as the MM system can consider every disease of mankind in every consultation.
The MDATA system includes a computer program that is designed to conduct a question and answer dialog with a human patient for the purpose of eliciting medical symptoms and then generating a list of possible illnesses suggested by the symptoms.
The MDATA system includes a comprehensive database of suitably formatted medical knowledge plus a driver program that conducts the interview, accesses the database, and interfaces with the user at run time.
The MDATA system is fully automated; no other human is involved at runtime other than the human user, who is assumed to be a patient or a patient representative such as a parent who is familiar with the patient/symptoms.
The MDATA system can, of course, also be used by medical students, nurses, even doctors who are willing to project themselves into the role of a patient. However, keeping in mind the purpose of the MDATA system, its medical terminology and style of questioning will tend to appear plodding, amateurish, and tedious to medical staff that, in recent decades, have lost the luxury of spending several hours at a time with a single patient.
The MDATA system output can be forwarded to a human MD, to another computer program for further use, and to the patient's medical record for continuing care such as follow-up or disease management.
Brief Review of Matrix System
A healthcare giver can quickly identify a disease using the Matrix system. This is done by entering chief complaints and other available information such as the cause and anatomic system.
The Matrix system provides a “population” specific differential diagnosis. Population specific refers to the age, sex, race, location, work place, travel history of the patient, etc. The Matrix system allows the user to look at disease information by an anatomic system that the disease may occur in, the possible causes of a problem, the chief complaint, by age groups, by gender, etc. The Matrix system allows the user to dynamically change the sorting parameters themselves. The Matrix system is fast, easy to use and structures the information in such a way as to be meaningful to a clinician at the bedside. The Matrix system can adjust to the sophistication of the user. For example, there are versions for Emergency Medical Technicians (EMTs), Nurses, Primary care physicians, Emergency Physicians, and Specialists.
Section I: Brief Review of Clinical Decision Making
There are four different levels of clinical decision making used in the diagnostic process. Each of these levels is represented or modeled by one or more of the MM system features.
The most basic level is pattern matching, sometimes called a “nearest neighbor” analysis. In this method, the PHIs (Patient Health Items) of a patient are simply matched to the PHIs of different diseases, and the one that most matches is the most likely diagnosis.
In certain embodiments, this is exactly where the MM system begins its diagnostic process. The initial differential diagnosis is assembled from the key words selected by the patient and the chief complaint set selected by the user. Each PHI is weighted differently in each disease, and, in the fully time-based system, weighted differently for each part of the disease time line.
The second level of clinical decision making is the “rule-based” method. In certain embodiments, rules refer to “if/then” statements, algorithms, or check lists. The algorithmic aspect of the MM system models this. So do the rules that are used by the diagnostic engine or list-based engine in dynamically sequencing questions. This process takes the following diagnostic aspects of the diseases into consideration: relativity, urgency, seriousness, and prevalence.
The third level of clinical decision is the “Hypothetical-Deductive Process”. The process of hypothesis generation, evaluation and refinement is modeled in the different axis of inquiry modes and a disease segregation feature.
The Axis of Inquiry (AOI) concept is where, in Horizontal Axis of Inquiry (HAI) mode, each disease object “votes” for what question it wants to ask to help clarify the disease. The list-based engine sequences the questions based upon several different diagnostic strategies. Each PHI is weighted differently in each disease, and as each question is asked and answered, the diagnostic weight of the PHIs, either positive or negative, is added. This continues until a threshold is reached, and one of the disease objects seems to be the most likely candidate. The threshold is related to the diagnostic score but also to how rapidly it is rising. At this point (threshold), the system goes into the Vertical Axis of Inquiry (VAI) mode.
In VAI, one disease object controls the sequence of the questions asked. The hypothesis is that the disease that is currently being evaluated is the diagnosis. If this assumption proves wrong, the system returns to the Horizontal Axis of Inquiry (HAI), in which other diseases are again considered in the differential diagnosis.
The fourth kind of clinical decision making is called the naturalistic or event-driven process. Here, certain situations (rather than diagnoses) are used to alter the diagnostic process. This frequently involves treating a patient's symptoms and signs before any diagnosis is known. This process can be seen in the Action Plateau, Nexus Point, Symptom Watch, and Treatment Finesse features, described below.
When a set of clinical PHIs suggest a situation that could require immediate attention, the Action Plateau (AP) function recognizes the situation and notifies the caregiver or patient.
There are some Meta heuristics built into the architecture of the MM system. For example, the system always considers the most urgent conditions first, even if their probability might delay their consideration. In one embodiment, the Treatment Finesse feature of the MM system always maximizes the therapeutic window of opportunity. In addition, the system weights all of the diseases in the differential diagnosis each time a PHI is established. That is referred to in the system as parallel processing or list-based processing.
A. Diagnostic Engine—Disease Segregation and Special PHI Designations
Several improvements have been made to the diagnostic system. A significant improvement is that the diagnostic system is now able to segregate diseases into several categories. This helps to speed the diagnostic process by reducing the number of questions that are asked of the patient without losing sensitivity. There are several ways disease segregation is implemented.
Firstly, when a disease can no longer reach diagnostic threshold, it is designated impossible to rule in at this time. Therefore, this disease is no longer considered in the differential diagnosis and is no longer allowed to suggest PHIs to the system.
Secondly, a typical diagnostic process would consider between 3 and 6 candidate diseases. Each of these diseases has a diagnostic score, and the rate of rise of diagnostic scores and number of hits make them most likely to be the diagnosis.
Thirdly, when a threshold is reached, the diseases that do not reach or exceed this threshold are “segregated” out of the diagnostic process. These segregated diseases are no longer able to “suggest” questions to be asked. Each time a PHI is established, however, if it occurs in the segregated diseases, it will be appropriately weighted. Further, diseases continuously being segregated out and then integrated back into the diagnostic process are most likely candidate diseases.
Certain PHIs now have special designations in a disease. They are as follows:
Besides the diagnostic weight, each PHI is given a “sine qua non” or “sine” score in each disease. This has to do with how important it is for the PHI to be present in order to have enough evidence to make a clinical diagnosis. The following are examples of the sine designations:
If a pathognomonic PHI is present, the disease is diagnosed. For example, seeing fortification figures before a headache is pathognomonic for migraine headache. That means that if the patient has this PHI then the patient has this disease because that PHI is 100% specific. Sine qua non—major means that that PHI or a specific number of PHIs out of a limited set of PHIs must be present for the diagnosis to be made.
The MDATA diagnostic engine can be referred to as the diagnostic engine or the list-based processor. The function of the engine is finding the next best question(s) to present to the user to be evaluated.
As discussed, diseases that cannot be ruled in are segregated out of the interaction and diseases that are lagging behind the main candidate disease are no longer allowed to suggest PHIs.
Once the VAI threshold is reached, the effective candidate disease set has been narrowed down to one. This is because in VAI, only one disease object proposes questions to be evaluated. However, all diseases in diagnostic consideration are still being weighted.
B. Features of the MM System
The MM system's main screens are originally composed into five parts:
Referring to an exemplary main screen display shown in
A case pane 106 contains the relevant clinical information about the patient. This comes either from the patient's medical record (if available) or the information acquired from the interaction with the patient or health care worker. This allows a health care worker to easily summarize the case for dictation or discussion with others.
A differential diagnosis pane 108 contains the list of possible diagnoses for the patient being evaluated. As each new piece of information or PHI is added to the system, the differential diagnosis is updated.
A menu bar 110 contains menu items that are further described in the section Menu Structure herein below.
In one format, the disease object window can be open at all times showing one disease object, or, more commonly, it is hidden from view until the heath care giver or patient go into the Vertical Axis of Inquiry (VAI)
An interview pane (not shown) presents questions directly to the user individually with a preamble that defines the question.
In addition, there are several “pop up” windows that are context dependent. These include the following:
The consultation tab sets the general parameters for the upcoming patient consultation. Variables for the consultation include the “sensitivity” level of the consultation, which describes how thorough the consultation will be. There are five levels of sensitivity.
1. The First Level of Sensitivity
At the first level (which is also the default level) all of the diseases encountered in emergency medicine and primary care are included in the run time database. For a disease to be included in the first level of sensitivity, it has a specified degree of at least one of three properties: urgency, seriousness, or high prevalence in the population from which the patient comes.
“Urgent” diseases are those in which an immediate diagnosis is critical for preserving the window of opportunity for therapeutic intervention. Examples of urgent diseases are acute coronary syndrome (ACS), myocardial infarction, or heart attack. These diseases are diagnosed as quickly as possible to allow for proper life-saving treatment.
“Serious” diseases have a longer time frame for diagnosis, therefore are considered serious but not urgent. An example of a serious but not urgent disease is a patient whose headache is caused by intra-cranial pressure, which in turn is caused by a brain tumor. This is to be diagnosed; however, the degree of urgency is less than that of a diagnosis of subarachnoid hemorrhage.
2. The Second Level of Sensitivity
A second level of sensitivity adds a set of disease defined generally by a medical specialty. Referring to
The column headings 200 represent anatomic systems of the body and the row headings 202 represent the different causes of diseases. These systems are used in most medical definitions.
In certain embodiments, the anatomic systems are as follows:
In certain embodiments, the causes of disease are as follows:
At the second level of sensitivity, the diseases that would typically be considered by a specialist are added to the run-time database from which the differential diagnosis is assembled. A subset of diseases can also be added to the differential diagnosis at any point in the evaluation of a patient. For example, if a patient has an eye problem, the user may add the ophthalmologic diagnoses considered by a general opthalmologist at the beginning of the consult.
3. The Third Level of Sensitivity
The third level of sensitivity considers diseases that focus on one specific segment of a specialty. An example would be retinal diseases of the eye.
4. The Fourth Level of Sensitivity
The fourth level of sensitivity includes those diseases that would be handled by a sub-specialty or a regional medical center focusing frequently on one chief complaint. Headache is an example of this sensitivity level.
5. The Fifth Level of Sensitivity
The fifth level of sensitivity considers all the diseases of mankind. This is variously calculated, but, in one embodiment, slightly more than 20,000 diseases are in the MM system database.
B. Sensitivity Factor Set
The consultation can be further customized by the use of the Sensitivity Factor Set (SFS) as defined in U.S. Pat. No. 5,660,176. In general, the SFS refers to the many variable thresholds in the systems which are used to establish the probability of diagnosis.
In one embodiment of the MM system, the SFS set is accompanied by a “slider” control consisting of an analog scale from 0 to 100. A setting of 100 would make it relatively “easy” for a disease to be considered and ruled in as a possible diagnosis.
This is accomplished by altering the scalar thresholds that are use to give different weights to different levels of a PHI. For example, various degrees of fever have a sliding scale of weights.
The more sensitive the system, the more new diagnostic possibilities are added to the differential list with each new PHI that is established. Generally, the differential diagnosis is assembled based on the initial input of PHIs including those selected by the patient as “key” or “clue” words, and the presenting chief complaint set selected by the user.
When a PHI is established that is found in a disease that is not in the differential list, and the sensitivity of the system in increased, it will be added to the differential. The more sensitive the system, the longer new disease possibilities will be added to the differential.
There is always a trade off between sensitivity and specificity, in regards to the amount of time to interact with the system. A more sensitive or “careful” system utilizes many more questions to be asked.
C. Scope of Inquiry
A third way of customizing the consultation is the use of the Scope of Inquiry (SOI) function. This refers to what the system does with pieces of information that do not match the disease that has been diagnosed. This is analogous to the review of systems (ROS) in a diagnostic evaluation performed by a doctor. The doctor will ask questions about each organ system of the body, even if not related to the patient's problem. If these symptoms are not found in the diagnosed disease, and if they are not found in the patients previously diagnosed diseases, they are “set aside” and dealt with at the end of the consultation. The SOI setting tells the MM system how aggressive to be with any unrelated complaints.
D. Time Context
A fourth way of customizing the MM system regards the use of time. The time context refers to the extent the MM system will consider time during the patient's consultation. A disease “declares” itself by how it unfolds through time, such as the sequence and patterns of the symptoms, the onset characteristics of each PHI and the way in which PHIs vary with time.
Thus, the use of time allows a disease to be more accurately diagnosed, and also to be diagnosed at the earliest point in time. Again, though, there is a tradeoff because of the time it takes to establish the absolute or relative onset times of all of the PHIs, how long each lasted, and the offset characteristics (e.g., sudden or gradual), etc.
E. Diagnostic Ranking Formulas
Referring to
The ranking formulas 904 are actually customizable by the user. In addition, the contribution of each of the aspects that are used in the ranking formula, for instance Percent of Score 900 and Urgency 902, can be changed.
Section III. Data Input
The data input system is designed to allow the user to enter in specific kinds of information with the fewest possible keystrokes or mouse clicks. Input is generally sequenced similar to a diagnostic to facilitate a smooth flow through the data input system.
A. Population Tab
To be accurate, differential diagnoses should be “population” specific. That is, basic pieces of demographic information can radically change the diseases to be considered in the upcoming evaluation. Referring to
The population tab allows the following information to be input:
There are certain subsets of patients that share features such that there is a difference in the diseases that one might expect to find for a given set of presenting symptoms. This tab allows the user to quickly indicate if a patient is in one of these subsets. Examples of special populations are as follows:
The time tabs seeks information about time context of the patient's illness. Specifically, the time tab looks to know whether the patient is still suffering from the complaints that prompted medical attention, or if the symptoms have gone away.
In addition, the duration of patient's illness is generally very important because many diseases can be excluded from diagnostic consideration due to this fact alone.
D. “Key Words” Tab
Referring to
E. Vital Sign Tab
Vital signs can be quickly entered in the vital sign tab. These can either be expressed as ranges or by the precise value. Most vital sign abnormalities (e.g. fever, hypotension) are also shown in the chief complaint set tab.
F. Chief Complaint Set Tab
Referring to
The chief complaints are linked to the “key” words. All diseases considered by MM are indexed by both the chief complaint and by the “key” words, and each word is weighted differently in each disease.
G. Matrix Tab
As discussed earlier, the anatomic-cause matrix, such as shown in
In certain embodiments, a single click will select a subset of diseases; while a double click causes that cell to “explode”. When a specific cell is pinpointed, that cell “explodes” or expands to reveal a nested two-dimensional array. In this display, the columns consist of subsets of the anatomic systems involved, and the rows consist of subsets of the causes of disease. To continue the example given above, the user may select a now nested cell defining viral disease of the bronchioles such as bronchiolitis in a child. The list of diseases added can be unfiltered by the PHIs already established at the time the matrix tab is invoked. The matrix tab may be used at any point in the consultation.
H. PHI Matrix Tab
Referring to
The mimics tab provides a detailed analysis of a selected disease as compared to other diseases that may have similar presentations. This feature allows a user to identify specific differences or wedge questions, which may help to distinguish one disease from other possible causes of the presenting PHIs.
Section IV: Work Up Progress
If the functions of the MM system are being used in the typical temporal sequence, by this time, the population specific data and the patient's presenting chief complaint set have been input and the system has established a population and a patient specific differential diagnosis.
To use the auditorium metaphor, at this point, we would have assembled a number of world-class disease specific specialists specifically to solve this case. It is here that each expert begins voting for what question he or she wants answered. The list-based processor here considers the diseases in deliberation, what is known about the patient and the relative urgency, seriousness and prevalence of the diseases in diagnostic consideration to dynamically assemble or sequence the questions.
In the MDATA system, and in the interview mode of the MM system, the questions are presented sequentially to the patient in simple language with a preamble which describes and defines the information the question is asking about.
In the MM system, these questions can be presented to the user in “batch” mode in a series of tabs called the Horizontal Axis of Inquiry tabs.
A. Horizontal Axis of Inquiry Tabs
The MDATA diagnostic engine runs in two modes, the first one being the horizontal axis of inquiry (HAI), and the second one being vertical axis of inquiry (VAI). In HAI, the most likely disease is not yet established, so questions pertaining to several diseases in the differential diagnosis are asked. When a patient indicates that a certain PHI is present, the disease specific diagnostic weight of that question is added to each respective disease object.
Each expert “listens” to all of the questions being asked. When the patient answers, the new information will make each disease either more likely, less likely, or its probability will not change.
Referring to
There are several “formulas” which are used to select the set of questions to be presented to the user or to select the one best question to the patient in interview mode. The formulas usually deal with the number of hits, misses, PHI weights, and PHI “sine” status, urgency, seriousness, and prevalence. As shown in
B. Case Pane
Referring again to
In certain embodiments, the case panel 106 shows the PHIs that the patient has (in green), the PHIs the patient has denied (in red), and the PHIs the patient is unsure of (in white). If the medical record is available, it can also contain a list of the patient's current diagnoses.
C. Differential Diagnosis Pane
Referring to pane 108 of
Continuing from left to right, in pane 108 of
The second phase of the diagnostic process is optional and many times not necessary. Frequently, as the care give reviews the differential diagnosis, it is possible to narrow the possibilities to five or fewer diseases. As mentioned briefly above, if the user elects to select the most likely candidates from the differential diagnosis and concentrate on those diseases, the user simply clicks in the “focus” column of diseases of interest in the differential diagnosis pane and clicks “segregate”. This will cause a sub-window to open containing only the selected diseases. At this point the HAI tabs will show questions taken from only the disease objects in the focused subset. As always, however, all diseases in the differential will continue to be weighted.
E. Sub-PHI Analysis
Each PHI in the system is a part of an inverted tree with the most general PHI being the root. For example, a tree may be as follows with the root being listed first: abdominal pain—right lower quadrant—colicky.
In addition, PHIs are modified and indexed by the N-dimensional array that constitutes a Timed Descriptor Spectra (TDS). These attributes include severity of pain, time shape of the onset, quality of pain, what precipitates it, and what makes it better. These help clinicians to more precisely define a symptom and are generally very important in diagnosis.
These attributes are shown to the user in the VAI mode in the disease object pane. But there are times when it is these attributes which allow a distinction to be made between several very similar disease processes. When the user has used the segregated disease or disease focus feature, these attributes are “visible” and are included in the HAI suggested questions. In addition to the TDS spectra attributes, a Gantt input pop up window is also available in the segregated disease analysis pane.
F. Disease Object Pane
Referring to
The disease object panel itself has a set of tabs nested inside it. The default view of the disease object is the one used by the clinician in the vertical axis of inquiry (VAI). The VAI is the mode of operation used when the diagnostic engine is focusing on a single diagnosis. At the same time, all of the PHIs that also occur in other disease objects are being weighted as they are established.
The default VAI tab shows all of the symptoms, signs and vital signs listed on the left side of the tab. In certain embodiments, all PHIs already established in the patient are shown in green, all PHIs denied are in red, and all unknown are in white. Therefore, the clinician can very easily see the PHIs that are to be evaluated to establish the diagnosis.
The value of each PHI and its associated weight is shown so that the clinician can generally concentrate on the most important PHIs first. There are several other sorting criteria that the clinician can use. Any PHI can be clicked on and it will open the PHI object. The PHI object shows all the diseases in which that PHI is present. This is a kind of instant differential diagnosis. Any disease may be selected and “opened” by clicking. In every disease object, in certain embodiments, the PHIs that match the disease are shown in green. Those that do not are shown in red. The weights of all of the PHIs in the disease object that is open are also displayed.
G. Disease Relationship Tab
The disease relationship tab displays a graphical representation of all of the risk factors, complications, and associated diseases that are related to the current disease. Risk factors and conditions that cause the disease are shown to the left. The differential diagnoses of this disease or the diseases that may present in a similar manner are listed directly under the diseases. Diseases that the current disease may cause or complications of this disease are shown to the right.
The relationships between the diseases are represented by a commonly used Java applet that allows the relationships to “fall into their natural position” after the graph is perturbed. (See The Visual Thesaurus® from Thinkmap, Inc. for example).
H. Disease Summary Tab
This tab contains a text summary of the disease.
I. Geographic Tab
The geographic tab has a world map in which the prevalence of the diseases is represented by different colors.
J. Single Disease Work Up Tab
A “work up” is defined as the tests (laboratory, imaging, and special tests like EKG) that are used in addition to history and physical examination to help establish the diagnosis.
In the Disease Object pane, the work up tab shows the current best laboratory test(s) of choice, the imaging modality of choice and special studies of choice to establish the diagnosis of that disease.
There are times when a clinician has limited the differential down to a small number of diseases by history and physical examination, but needs the better diagnostic resolution provided by addition testing.
In this case, the user can invoke the global work up function. This functions as follows. The clinician selects several possible diseases from the differential diagnosis and then selects the global workup function. This function retrieves the lab tests of choice, imaging modality of choice and special studies of choice for the disease selected. Then it deletes duplicates and displays the results to the user.
K. Disease Object Feedback Tab
By choosing the disease object feedback tab, the user can give their input on what disease object information may need modification, additions or corrections.
L. Disease—PHI “Surfing”
Again, in the disease object, a left click on any PHI will instantly open the PHI object in which all of the diseases that contain that PHI are listed. Any disease may be selected by clicking and that disease object is opened. The PHIs that the patient has are colored green and the PHIs that do not match the disease are colored red.
M. Central Feedback
The MM system can be linked to a central web site to exchange data about possible WMD (Weapons of Mass Destruction) or possible epidemics. This link can also facilitate the exchange of data regarding new treatments of choice or information regarding new tests of choice.
Section V: The Interview Mode
The MM system can be run in a fully automated mode, which in the MM system is referred to as the interview mode (IM), such as shown in
There are three differences in the way in which questions are presented in the IM of the MM system. First, the questions are presented one at a time, sequenced by the list-based or diagnostic engine based on the “votes” by the disease objects. Second, each question is preceded by a preamble or introduction that precisely defines the information that is being requested of the patient. Finally, the language used is very basic and is intended for someone with no medical background. Specific aspects of the questions, like the language used, the sophistication of the language can be controlled by the “Question Roller” function discussed in U.S. Pat. No. 6,234,964, which is hereby incorporated by reference.
In IM, the patient may be requested to perform a physical self-examination, which may call for the assistance of another person. This person does not need to be medically trained to assist in the examination, as the IM will list step-by-step instructions in non-medical terms. Thus, the provisional diagnosis is based solely on a patient's medical history and physical self-examination.
Section VI: other Pop-Up Data Entry Windows
A. Gantt Input Control
As previously mentioned, in general, one of the most important factors in determining a correct diagnosis may be to consider the timing of the onset of the patient's symptoms. A Gantt chart is a horizontal bar chart developed for use as a production control tool. In certain embodiments, the MM system contains a Gantt chart which displays all of the patient's symptoms in order of appearance. These symptoms are displayed as horizontal bars which can be rearranged to match the onset sequence of the patient's illness. This information is then interpreted by the MM system and the diseases that most closely match this pattern are given extra diagnostic weight.
B. Graphic Input of Onset and Other Time Shapes
Generally, the onset “time shape” of a symptom may be very important. For example, the sudden onset of a pain suggests a vascular or colic event. A discontinuous onset (i.e., a symptom that occurs and then stops then reoccurs) is also frequently a vascular problem. For example, angina pectoris, pain from the heart caused by insufficient oxygenated blood, often “stutters” in its onset.
Onset types include exponential onset, linear, and sinusoidal. Each has clinical significance.
In order to quickly indicate the onset characteristics of a PHI, a simple graphic input window opens and the user or patient may indicate the time shape of the onset of the symptoms by the way they vary through time and their offset. A Runge curve fitting algorithm is used to transform the shape input for the user to several suggested forms.
Several other time shapes are generally important in diagnosis. For example, when the severity of pain through time plots a sinusoidal curve, frequently a cylindrically shaped structure is obstructed. Renal colic, biliary colic, crampy abdominal pain of small bowel obstruction and pain of labor are examples.
Onset and offset analysis is accomplished by another input popup window which plots the severity of a symptom through time. Analysis of offset time shapes and onset of symmetry are supported as well as all of the patient's symptoms considered together.
C. Nexus Point Action Plateau and Symptom Watch Window
In the MDATA system, the Nexus Point (NP) is defined as that time in a patient's illness when the symptoms seemed to have “turned the corner” and begin to improve. Generally, this is an important point because, frequently, if the patient is improving on their own, further workup is deferred to allow “nature to affect the cure” as Sir William Osler recommends.
There is a small window that contains “meta” information about the consultation. Included in this input area is a place for the patient to indicate if he believes he may be beginning to improve.
The MDATA system also provides an “Action Plateau” (AP) feature which warns the patient that a dangerous situation has been reached, and even taking the time to continue the current consultation could be detrimental. The same function is present in the MM system. The use of a simple graphic “risk thermometer” warns the patient and/or physician of a potentially dangerous situation for the patient, the care giver, or both.
During a consultation, particularly in a fully time-based one, the appearance of certain symptoms can change the actions that are to be taken. The example of appendicitis will be used to explain the principle. A patient that has anorexia, and poorly localized epigastric abdominal pain with nausea and/or vomiting may have early appendicitis. In appendicitis, the Diagnostic Point Symptom (DPS) is the movement of this pain into the right lower quadrant of the abdomen. When this occurs, there is a significant increase in the probability that the patient does have appendicitis, even before RLQ tenderness or rebound tenderness develops. The Symptom Watch (SW) function allows immediate action to be taken as soon as the DPS occurs. The SW window has a list of symptoms that the patient is instructed to click on if any of these symptoms occur during the interview mode.
D. Alternative Input Tabs
Typically, the patient describes several symptoms which are used to retrieve diseases to formulate the differential diagnosis. These disease objects are then used as the basis for enquiring about other symptoms, signs, etc.
There is an alternative way for the patient or caregiver to enter data into the system. The evaluation may be approached from a traditional doctor-patient interaction through a series of nested outlines or matrices.
The MM system can follow the traditional input sequence of information allowing the user to first enter the chief complaint set, history of the present illness, review of systems, past medical history, family medical history, social history, vital signs and physical examination findings. In the same way that each PHI is really an inverted tree, in certain embodiments, so are the other PHIs structured and presented to the user. For example, each symptom starts from the anatomic system involved. Then the past medical history, family history and social history are structured. The same is true for physical examination maneuvers.
E. All Slider Tabs
The process of varying one parameter in a clinical case can be very helpful in the diagnostic process. For example, if the age of patient is varied, the differential diagnosis will change depending upon the age. This is somewhat analogous to a “sensitivity analysis” in accounting. The slider controls allow the following parameters to be dynamically changed
Again, in the same way that a user can look up the tests utilized to confirm one diagnosis in a disease object's work up tab, there are times when several diagnostic possibilities are still under serious consideration and a testing plan is to be created to distinguish between the various possibilities.
The multiple disease work up tab accomplishes this. Any set of possible diagnoses can be selected, and the work up function invoked, which will cause a testing plan (usually expressed as an algorithm) to be displayed in the work up tab. This is most frequently used when a final set of diagnoses is being considered in the segregated or focused disease set.
G. Panic Tab
There are certain standardized ways of handling very critical situations in medicine. The most easily understood examples are Advanced Cardiac Life Support (ACLS) and Cardio Pulmonary Resuscitation (CPR). In each case, very specific steps, procedures, and medications are indicated. They are usually committed to memory because there is not sufficient time to look up information in these time-critical situations.
The Panic tab has instructions for the Heimlich Hug, CPR and ACLS available with one or two clicks in the MM system. The contents are actually kept in the treatment tables so they can be updated as often as necessary.
H. Reports Tab
Due to the fact that all questions asked and answered have a time and date stamp, the system can prepare a detailed report for both patients and caregivers. The standard MDATA report contains the following information:
As in the MDATA system, the users are in partnership with the entity maintaining the system to constantly improve and expand the system. The MM system is a constantly evolving system. Generally, an important part of this involves the update and feedback loops built into the system. As new tests, imaging modalities and treatments become available, the MM system is updated when it is connected to the website of the entity maintaining the system.
In certain embodiments, the following aspects of the MM system are updated:
In addition to downloading of information, important feedback information to monitor and improve the system is uploaded whenever the MM system application connects to the website of the entity maintaining the system.
Feedback information from the MM systems to the website includes the following information:
It is often said that there is no substitute for experience. Expert clinicians refine their diagnostic skill over time by comparing their patients' symptoms to both what they thought was the diagnosis to the “final” diagnosis. This process is not only reproducible but the experience of many clinicians can be pooled, and what is learned quickly translated into better diagnostic weights and a more accurate and useful system.
In order to accomplish this, a system based on continuous feedback, analysis and modification has been created. User feedback allows the entity maintaining the system to alter such parameters as the diagnostic weights through time to be more accurate. Currently, there are no universally accepted diagnostic weights of PHIs. This is particularly true when the diagnostic weight of a PHI changes depending upon the disease timeline.
The MM system is based primarily on history, physical examination findings, and vital signs, such as pulse and temperature. A final diagnosis is usually only possible with an imaging modality (x-ray of a broken bone, CT scan of a sub-arachnoid hemorrhage or a lab test, such as T3, T4 and TSH in hyperthyroidism.) This is why it is generally so important to indicate exactly how the final diagnosis is ascertained.
The idea is to compare what the MM system thought the diagnosis was to what turned out to be the final diagnosis (to the extent it is known under the circumstances of the case.) The user of the system then can designate what the final diagnosis is for each consultation. This is accomplished by completing a check list and a “slider” control, both of which are used to estimate the probability that the “final” diagnosis is accurate.
Diagnoses are considered “final” based on various criteria that are specific to each disease. This information is found in the disease object. Once the clinician enters the final diagnosis into the system, the information is uncoupled from any patient identification information, and sent to the website for analysis. When the information is at the website, the PHIs that were present and their values are compared to the current set of diagnostic weights. This information is then statistically pooled and used to continuously update the weight placed on each PHI.
C. Data Siphon
Data siphon is the function in which the statistical data about the consultation is decoupled from the patient identification information. Each clinical interaction is given a unique code; however, no patient identification information is included in order to protect the patient's right to privacy. Generally, patient anonymity is to be respected. None of the statistical information about diseases can be traced back to a patient's medical record.
D. The Inverse Relationship of Sensitivity and Specificity
There is an inverse relationship between the sensitivity and the specificity of a system. A system with 100% sensitivity (meaning never missing a diagnosis) would have to consider every disease in every patient since any disease could be in the very early phase. A system with 100% selectivity (meaning never diagnosing a disease that was not present) could only achieve this by never diagnosing any disease. Therefore, a balance is set to achieve a balance between these two parameters. The SFS, SOI, and sensitivity “level” are examples of having a “tunable” system.
E. Multiple Stage Diagnosis
In the MM system, the diagnostic process is divided into several stages so that a provisional or clinical diagnosis can be established before testing begins. As stated above, it may not be possible to ascertain the “final” diagnosis until autopsy.
In certain embodiments, the MM system arrives at a provisional diagnosis based on the patient's history, physical examination and available vital signs. After testing is complete, the test results can be added into the MM system. Each test, like each PHI, is weighted differently in each disease. This allows the MM system to significantly increase the probability of making the “final” diagnosis. In addition, this information is generally very important to the feedback system.
F. Other Features
There are several other MDATA features that are incorporated into the MM system. Briefly, they are as follows:
Health Care Giver Mode assumes that a caregiver will involved in asking the questions and examining the patient. This mode functions best when there are testing facilities available but generally does not require them.
B. Parsimony Mode
The MM system's default method of diagnosis is based upon the Law of Parsimony. In medical terms, this means that all of the patient's symptoms are being caused by one new disease process.
An Existing Disease Subtraction Layer (EDSL) is used to “subtract” away the patient's underlying complaints due to any other chronic disease(s) in order to establish the efficacy of therapy of the disease being managed. This means that even if some of the patient's symptoms are caused by underlying disease process, the patient is usually able to differentiate different “kinds” of the same symptom. For example, a patient with chronic obstructive pulmonary disease may experience an underlying dyspnea or shortness of breath. This is frequently precipitated by exertion. A patient with left sided congestive heart failure also becomes short of breath but usually when laying flat. This is called orthopnea. It is typically easy for the patient to make this distinction if the correct questions are asked.
In the MM system, the reliance on efficacy of the law of parsimony can be increased by using the EDSL and a carefully constructed and worded history.
C. Negative Weights
The use of “negative” weights can be very helpful in the diagnostic process, but are most valuable when it is assumed that all of the patent's current PHIs are caused by one new pathologic process. Thus they can be problematic, particularly when the patient has underlying concurrent chronic disease processes. The “Existing Disease Subtraction Layer” (EDSL) addresses this issue, by allowing the PHIs of existing diseases to be subtracted from the case. By assuming parsimony, the more PHIs a patient has that do not appear in a given disease, the less likely that disease is to be the diagnosis. The use of negative weights has a built in “synergy” in which the sum of the weight subtracted from each disease is further increased based on the number of non-matching PHIs present.
Another use of negative synergy is when two PHIs actually measure the same pathologic process. For example, in left sided congestive heart failure, paroxysmal nocturnal dyspnea and orthopnea reflect rising pulmonary artery pressure. Thus the combined weight of having both these PHIs present would actually be less than the sum of their diagnostic weights.
In the fully time-based system, not having a PHI that does not occur until the later stages of a disease would tend to increase the probability that the patient is in an early phase of that disease process. For example, not having RLQ abdominal pain makes early appendicitis more likely.
Section IX: Menu Structure
In certain embodiments of the MM system, the main menu structure resembles the standard windows style conventions and consists of a series of nested menus.
The top tier menu structure is as follows:
For each item listed above, the following information is the second tier:
Referring to
Referring to
Referring again to
Referring to
Process 1304 continues at state 1508 where a time context is selected. The time context specifies if this disease process is current and ongoing, in the past, or a hypothetical situation about the possible future. Moving to state 1510, the facilities available are selected. This allows the system to intuitively ask for more advanced findings, only if they are available, such as: EKG reading, X-ray findings, or CAT scan results. Moving to state 1512 the user is asked to specify the number of questions or PHIs they would like presented at one time. This can be adjusted from interview mode, where a single question is presented directly to the patient, or it can be set to display questions and PHIs in batch mode, allowing faster input of responses by dealing with them as a group. Proceeding to state 1514, the remaining user/consultation options are specified. These include options such as display setting, font usage, and other general options. At the completion of state 1514, process 1304 advances to return state 1516 and control is then returned to
Referring again to
Referring to
Referring again to
Moving to state 1310, the user is asked to select from a list of chief complaints that might apply to this consultation. Chief complaints are medically significant signs or symptoms that are used by the system to create its initial differential diagnosis. Advancing to state 1312, the system utilizes all of the information gathered from the user thus far to generate an initial population specific differential diagnosis. This serves as a starting point for the diagnostic process. Process 1300 then advances to main loop process 1314, which is further described in conjunction with
Referring to
Integration/segregation process 1704 is described in
Referring again to
Referring again to
Referring to
Referring to
Referring again to
Referring to
Referring to
Referring back to
Referring again to
Referring back to
Referring to
Referring back to
Referring back to
Referring to
Referring back to
Conclusion
Specific blocks, sections, devices, functions, processes and modules may have been set forth. However, a skilled technologist will realize that there are many ways to partition the system, and that there are many parts, components, processes, modules or functions that may be substituted for those listed above.
While the above detailed description has shown, described and pointed out the fundamental novel features of the invention as applied to various embodiments, it will be understood that various omissions and substitutions and changes in the form and details of the system illustrated may be made by those skilled in the art, without departing from the intent of the invention. The foregoing description details certain embodiments of the invention. It will be appreciated, however, that no matter how detailed the foregoing appears, the invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiment is to be considered in all respects only as illustrative and not restrictive and the scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.
The benefit under 35 U.S.C. §119(e) of U.S. Provisional Application No. 60/621,384, filed Oct. 22, 2004, entitled “MATRIX-MDATA SYSTEM AND METHOD” is hereby claimed, and this application is incorporated by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3970996 | Yasaka et al. | Jul 1976 | A |
| 4051522 | Healy | Sep 1977 | A |
| 4220160 | Kimball et al. | Sep 1980 | A |
| 4290114 | Sinay | Sep 1981 | A |
| 4315309 | Coli | Feb 1982 | A |
| 4337377 | Van Riper et al. | Jun 1982 | A |
| 4428381 | Hepp | Jan 1984 | A |
| 4458693 | Badzinski et al. | Jul 1984 | A |
| 4465077 | Schneider | Aug 1984 | A |
| 4531527 | Reinhold, Jr. et al. | Jul 1985 | A |
| 4606352 | Geddes et al. | Aug 1986 | A |
| 4712562 | Ohayon et al. | Dec 1987 | A |
| 4731726 | Allen, III | Mar 1988 | A |
| 4733354 | Potter et al. | Mar 1988 | A |
| 4770189 | Shyu | Sep 1988 | A |
| 4803625 | Fu et al. | Feb 1989 | A |
| 4825869 | Sasmor et al. | May 1989 | A |
| 4838275 | Lee | Jun 1989 | A |
| 4839822 | Dormond et al. | Jun 1989 | A |
| 4858121 | Barber et al. | Aug 1989 | A |
| 4868763 | Masui et al. | Sep 1989 | A |
| 4933873 | Kaufman et al. | Jun 1990 | A |
| 4945476 | Bodick et al. | Jul 1990 | A |
| 4962491 | Schaeffer | Oct 1990 | A |
| 4974607 | Miwa | Dec 1990 | A |
| 4975840 | DeTore et al. | Dec 1990 | A |
| 5012411 | Policastro et al. | Apr 1991 | A |
| 5012815 | Bennett et al. | May 1991 | A |
| 5023785 | Adrion et al. | Jun 1991 | A |
| 5030948 | Rush | Jul 1991 | A |
| 5054493 | Cohn et al. | Oct 1991 | A |
| 5084819 | Dewey et al. | Jan 1992 | A |
| 5099424 | Schneiderman | Mar 1992 | A |
| 5113869 | Nappholz et al. | May 1992 | A |
| 5193541 | Hatsuwi | Mar 1993 | A |
| 5196682 | Englehardt | Mar 1993 | A |
| 5228449 | Christ et al. | Jul 1993 | A |
| 5235510 | Yamada et al. | Aug 1993 | A |
| 5241621 | Smart | Aug 1993 | A |
| 5255187 | Sorensen | Oct 1993 | A |
| 5257627 | Rapoport | Nov 1993 | A |
| 5263123 | Hayashi | Nov 1993 | A |
| 5265613 | Feldman et al. | Nov 1993 | A |
| 5299121 | Brill et al. | Mar 1994 | A |
| 5307263 | Brown | Apr 1994 | A |
| 5337752 | Reeves | Aug 1994 | A |
| 5347632 | Filepp et al. | Sep 1994 | A |
| 5357427 | Langen et al. | Oct 1994 | A |
| 5377258 | Bro | Dec 1994 | A |
| 5390238 | Kirk et al. | Feb 1995 | A |
| 5404292 | Hendrickson | Apr 1995 | A |
| 5415167 | Wilk | May 1995 | A |
| 5418888 | Alden | May 1995 | A |
| 5421343 | Feng | Jun 1995 | A |
| 5435324 | Brill | Jul 1995 | A |
| 5437278 | Wilk | Aug 1995 | A |
| 5441047 | David et al. | Aug 1995 | A |
| 5442728 | Kaufman et al. | Aug 1995 | A |
| 5463548 | Asada et al. | Oct 1995 | A |
| 5471382 | Tallman et al. | Nov 1995 | A |
| 5473537 | Glazer et al. | Dec 1995 | A |
| 5481647 | Brody et al. | Jan 1996 | A |
| 5486999 | Mebane | Jan 1996 | A |
| 5517405 | McAndrew et al. | May 1996 | A |
| 5519433 | Lappington et al. | May 1996 | A |
| 5533522 | Feng | Jul 1996 | A |
| 5541977 | Hodges et al. | Jul 1996 | A |
| 5544649 | David et al. | Aug 1996 | A |
| 5553609 | Chen et al. | Sep 1996 | A |
| 5555169 | Namba et al. | Sep 1996 | A |
| 5572421 | Altman et al. | Nov 1996 | A |
| 5583758 | McIlroy et al. | Dec 1996 | A |
| 5594638 | Iliff | Jan 1997 | A |
| 5596994 | Bro | Jan 1997 | A |
| 5601435 | Quy | Feb 1997 | A |
| 5619991 | Sloane | Apr 1997 | A |
| 5622171 | Asada et al. | Apr 1997 | A |
| 5633910 | Cohen | May 1997 | A |
| 5642731 | Kehr | Jul 1997 | A |
| 5642936 | Evans | Jul 1997 | A |
| 5659793 | Escobar et al. | Aug 1997 | A |
| 5660176 | Iliff | Aug 1997 | A |
| 5672154 | Sillén et al. | Sep 1997 | A |
| 5675760 | Houwen et al. | Oct 1997 | A |
| 5678562 | Sellers | Oct 1997 | A |
| 5692220 | Diamond et al. | Nov 1997 | A |
| 5692501 | Minturn | Dec 1997 | A |
| 5694939 | Cowings | Dec 1997 | A |
| 5703786 | Conkright | Dec 1997 | A |
| 5711297 | Iliff | Jan 1998 | A |
| 5722418 | Bro | Mar 1998 | A |
| 5724968 | Iliff | Mar 1998 | A |
| 5724983 | Selker et al. | Mar 1998 | A |
| 5732397 | DeTore et al. | Mar 1998 | A |
| 5746204 | Schauss | May 1998 | A |
| 5769074 | Barnhill et al. | Jun 1998 | A |
| 5772585 | Lavin et al. | Jun 1998 | A |
| 5778882 | Raymond et al. | Jul 1998 | A |
| 5788640 | Peters | Aug 1998 | A |
| 5794208 | Goltra | Aug 1998 | A |
| 5800347 | Skates et al. | Sep 1998 | A |
| 5802495 | Goltra | Sep 1998 | A |
| 5812984 | Goltra | Sep 1998 | A |
| 5822715 | Worthington et al. | Oct 1998 | A |
| 5823949 | Goltra | Oct 1998 | A |
| 5828943 | Brown | Oct 1998 | A |
| 5832448 | Brown | Nov 1998 | A |
| 5835900 | Fagg, III et al. | Nov 1998 | A |
| 5839430 | Cama | Nov 1998 | A |
| 5862304 | Ravdin et al. | Jan 1999 | A |
| 5868669 | Iliff | Feb 1999 | A |
| 5879163 | Brown et al. | Mar 1999 | A |
| 5887133 | Brown et al. | Mar 1999 | A |
| 5897493 | Brown | Apr 1999 | A |
| 5899855 | Brown | May 1999 | A |
| 5908383 | Brynjestad | Jun 1999 | A |
| 5909679 | Hall | Jun 1999 | A |
| 5910107 | Iliff | Jun 1999 | A |
| 5911132 | Sloane | Jun 1999 | A |
| 5918603 | Brown | Jul 1999 | A |
| 5933136 | Brown | Aug 1999 | A |
| 5935060 | Iliff | Aug 1999 | A |
| 5940801 | Brown | Aug 1999 | A |
| 5951300 | Brown | Sep 1999 | A |
| 5953704 | McIlroy et al. | Sep 1999 | A |
| 5956501 | Brown | Sep 1999 | A |
| 5960403 | Brown | Sep 1999 | A |
| 5974389 | Clark et al. | Oct 1999 | A |
| 5976082 | Wong et al. | Nov 1999 | A |
| 5987519 | Peifer | Nov 1999 | A |
| 5997476 | Brown | Dec 1999 | A |
| 6001060 | Churchill et al. | Dec 1999 | A |
| 6022315 | Iliff | Feb 2000 | A |
| 6032119 | Brown et al. | Feb 2000 | A |
| 6032678 | Rottem | Mar 2000 | A |
| 6071236 | Iliff | Jun 2000 | A |
| 6081786 | Barry et al. | Jun 2000 | A |
| 6101478 | Brown | Aug 2000 | A |
| 6106459 | Clawson | Aug 2000 | A |
| 6113540 | Iliff | Sep 2000 | A |
| 6117073 | Jones et al. | Sep 2000 | A |
| 6144837 | Quy | Nov 2000 | A |
| 6149585 | Gray | Nov 2000 | A |
| 6151586 | Brown | Nov 2000 | A |
| 6161095 | Brown | Dec 2000 | A |
| 6167362 | Brown | Dec 2000 | A |
| 6168563 | Brown | Jan 2001 | B1 |
| 6177940 | Bond et al. | Jan 2001 | B1 |
| 6186145 | Brown | Feb 2001 | B1 |
| 6189029 | Fuerst | Feb 2001 | B1 |
| 6196970 | Brown | Mar 2001 | B1 |
| 6206829 | Iliff | Mar 2001 | B1 |
| 6230142 | Benigno et al. | May 2001 | B1 |
| 6233539 | Brown | May 2001 | B1 |
| 6234964 | Iliff | May 2001 | B1 |
| 6240393 | Brown | May 2001 | B1 |
| 6246975 | Rivonelli et al. | Jun 2001 | B1 |
| 6246992 | Brown | Jun 2001 | B1 |
| 6247002 | Steels | Jun 2001 | B1 |
| 6248063 | Barnhill et al. | Jun 2001 | B1 |
| 6248065 | Brown | Jun 2001 | B1 |
| 6260022 | Brown | Jul 2001 | B1 |
| 6270455 | Brown | Aug 2001 | B1 |
| 6270456 | Iliff | Aug 2001 | B1 |
| 6330426 | Brown et al. | Dec 2001 | B2 |
| 6334778 | Brown | Jan 2002 | B1 |
| 6368273 | Brown | Apr 2002 | B1 |
| 6375469 | Brown | Apr 2002 | B1 |
| 6379301 | Worthington et al. | Apr 2002 | B1 |
| 6381577 | Brown | Apr 2002 | B1 |
| 6383135 | Chikovani et al. | May 2002 | B1 |
| 6468210 | Iliff | Oct 2002 | B1 |
| 6475143 | Iliff | Nov 2002 | B2 |
| 6482156 | Iliff | Nov 2002 | B2 |
| 6524241 | Iliff | Feb 2003 | B2 |
| 6527713 | Iliff | Mar 2003 | B2 |
| 6569093 | Iliff | May 2003 | B2 |
| 6597392 | Jenkins et al. | Jul 2003 | B1 |
| 6598035 | Branson et al. | Jul 2003 | B2 |
| 6601055 | Roberts | Jul 2003 | B1 |
| 6641532 | Iliff | Nov 2003 | B2 |
| 6687685 | Sadeghi et al. | Feb 2004 | B1 |
| 6725209 | Iliff | Apr 2004 | B1 |
| 6730027 | Iliff | May 2004 | B2 |
| 6736776 | Miles | May 2004 | B2 |
| 6746399 | Iliff | Jun 2004 | B2 |
| 6748353 | Iliff | Jun 2004 | B1 |
| 6764447 | Iliff | Jul 2004 | B2 |
| 6767325 | Iliff | Jul 2004 | B2 |
| 6770029 | Iliff | Aug 2004 | B2 |
| 6817980 | Iliff | Nov 2004 | B2 |
| 6849045 | Iliff | Feb 2005 | B2 |
| 6900807 | Liongosari et al. | May 2005 | B1 |
| 6903657 | Kwoen | Jun 2005 | B2 |
| 6968375 | Brown | Nov 2005 | B1 |
| 7076437 | Levy | Jul 2006 | B1 |
| 7081091 | Merrett et al. | Jul 2006 | B2 |
| 7149756 | Schmitt et al. | Dec 2006 | B1 |
| 7167818 | Brown | Jan 2007 | B2 |
| 7223235 | Brown | May 2007 | B2 |
| 7223236 | Brown | May 2007 | B2 |
| 7252636 | Brown | Aug 2007 | B2 |
| 7258666 | Brown | Aug 2007 | B2 |
| 7259681 | Kwoen | Aug 2007 | B2 |
| 7264591 | Brown | Sep 2007 | B2 |
| 7297108 | Iliff | Nov 2007 | B2 |
| 7297109 | Brown | Nov 2007 | B2 |
| 7297111 | Iliff | Nov 2007 | B2 |
| 7300402 | Iliff | Nov 2007 | B2 |
| 7305348 | Brown | Dec 2007 | B1 |
| 7306560 | Iliff | Dec 2007 | B2 |
| 7310668 | Brown | Dec 2007 | B2 |
| 7320030 | Brown | Jan 2008 | B2 |
| 7344496 | Iliff | Mar 2008 | B2 |
| 7392167 | Brown | Jun 2008 | B2 |
| 7399276 | Brown et al. | Jul 2008 | B1 |
| 7421140 | Rottem | Sep 2008 | B2 |
| 7516192 | Brown | Apr 2009 | B2 |
| 7698154 | Marchosky | Apr 2010 | B2 |
| 7769600 | Iliff | Aug 2010 | B2 |
| 7780595 | Iliff | Aug 2010 | B2 |
| 7899687 | Morris | Mar 2011 | B2 |
| 20010012913 | Iliff | Aug 2001 | A1 |
| 20010039503 | Chan et al. | Nov 2001 | A1 |
| 20010053875 | Iliff | Dec 2001 | A1 |
| 20020019749 | Becker et al. | Feb 2002 | A1 |
| 20020029157 | Marchosky | Mar 2002 | A1 |
| 20020068857 | Iliff | Jun 2002 | A1 |
| 20020148477 | Kwoen | Oct 2002 | A1 |
| 20020170565 | Walker et al. | Nov 2002 | A1 |
| 20020186818 | Arnaud et al. | Dec 2002 | A1 |
| 20030036686 | Iliff | Feb 2003 | A1 |
| 20030069753 | Brown | Apr 2003 | A1 |
| 20030153819 | Iliff | Aug 2003 | A1 |
| 20030163299 | Iliff | Aug 2003 | A1 |
| 20030181790 | David et al. | Sep 2003 | A1 |
| 20030225315 | Merrett et al. | Dec 2003 | A1 |
| 20040017475 | Akers et al. | Jan 2004 | A1 |
| 20040019259 | Brown | Jan 2004 | A1 |
| 20040059200 | Iliff | Mar 2004 | A1 |
| 20040107116 | Brown | Jun 2004 | A1 |
| 20040116780 | Brown | Jun 2004 | A1 |
| 20040117207 | Brown | Jun 2004 | A1 |
| 20040117208 | Brown | Jun 2004 | A1 |
| 20040117209 | Brown | Jun 2004 | A1 |
| 20040117210 | Brown | Jun 2004 | A1 |
| 20040193377 | Brown | Sep 2004 | A1 |
| 20040199332 | Iliff | Oct 2004 | A1 |
| 20040199409 | Brown | Oct 2004 | A1 |
| 20040219500 | Brown | Nov 2004 | A1 |
| 20040249778 | Iliff | Dec 2004 | A1 |
| 20050010088 | Iliff | Jan 2005 | A1 |
| 20050010444 | Iliff | Jan 2005 | A1 |
| 20050027562 | Brown | Feb 2005 | A1 |
| 20050060194 | Brown | Mar 2005 | A1 |
| 20050080652 | Brown | Apr 2005 | A1 |
| 20050086083 | Brown | Apr 2005 | A1 |
| 20050134609 | Yu | Jun 2005 | A1 |
| 20050177391 | Shimizu et al. | Aug 2005 | A1 |
| 20050228883 | Brown | Oct 2005 | A1 |
| 20050234306 | Schulte et al. | Oct 2005 | A1 |
| 20050256739 | Brown | Nov 2005 | A1 |
| 20050273359 | Young | Dec 2005 | A1 |
| 20050273509 | Brown | Dec 2005 | A1 |
| 20060004611 | Brown | Jan 2006 | A1 |
| 20060010014 | Brown | Jan 2006 | A1 |
| 20060089969 | Brown | Apr 2006 | A1 |
| 20060100910 | Brown | May 2006 | A1 |
| 20060178914 | Brown | Aug 2006 | A1 |
| 20060189853 | Brown | Aug 2006 | A1 |
| 20060206012 | Merrett et al. | Sep 2006 | A1 |
| 20060234202 | Brown | Oct 2006 | A1 |
| 20060235722 | Brown | Oct 2006 | A1 |
| 20060241975 | Brown | Oct 2006 | A1 |
| 20060247951 | Brown | Nov 2006 | A1 |
| 20060247979 | Brown | Nov 2006 | A1 |
| 20060252089 | Brown | Nov 2006 | A1 |
| 20060253303 | Brown | Nov 2006 | A1 |
| 20060253574 | Brown | Nov 2006 | A1 |
| 20060253576 | Brown | Nov 2006 | A1 |
| 20060271404 | Brown | Nov 2006 | A1 |
| 20060285736 | Brown | Dec 2006 | A1 |
| 20060287889 | Brown | Dec 2006 | A1 |
| 20060287931 | Brown | Dec 2006 | A1 |
| 20060294233 | Brown | Dec 2006 | A1 |
| 20070016445 | Brown | Jan 2007 | A1 |
| 20070016446 | Brown | Jan 2007 | A1 |
| 20070016447 | Brown | Jan 2007 | A1 |
| 20070016448 | Brown | Jan 2007 | A1 |
| 20070021984 | Brown | Jan 2007 | A1 |
| 20070048691 | Brown | Mar 2007 | A1 |
| 20070055486 | Brown | Mar 2007 | A1 |
| 20070061167 | Brown | Mar 2007 | A1 |
| 20070067251 | Brown | Mar 2007 | A1 |
| 20070078681 | Brown | Apr 2007 | A1 |
| 20070094049 | Brown | Apr 2007 | A1 |
| 20070100665 | Brown | May 2007 | A1 |
| 20070100934 | Brown | May 2007 | A1 |
| 20070111176 | Brown | May 2007 | A1 |
| 20070118403 | Brown | May 2007 | A1 |
| 20070118404 | Brown | May 2007 | A1 |
| 20070124179 | Brown | May 2007 | A1 |
| 20070168226 | Brown | Jul 2007 | A1 |
| 20070168242 | Brown | Jul 2007 | A1 |
| 20070212671 | Brown | Sep 2007 | A1 |
| 20070213608 | Brown | Sep 2007 | A1 |
| 20070239490 | Sullivan | Oct 2007 | A1 |
| 20070299321 | Brown | Dec 2007 | A1 |
| 20080004915 | Brown | Jan 2008 | A1 |
| 20080045811 | Iliff | Feb 2008 | A1 |
| 20080046268 | Brown | Feb 2008 | A1 |
| 20080051638 | Iliff | Feb 2008 | A1 |
| 20080051639 | Iliff | Feb 2008 | A1 |
| 20080051640 | Iliff | Feb 2008 | A1 |
| 20080051641 | Iliff | Feb 2008 | A1 |
| 20080052116 | Iliff | Feb 2008 | A1 |
| 20080052118 | Iliff | Feb 2008 | A1 |
| 20080052119 | Iliff | Feb 2008 | A1 |
| 20080052120 | Iliff | Feb 2008 | A1 |
| 20080052121 | Iliff | Feb 2008 | A1 |
| 20080052122 | Iliff | Feb 2008 | A1 |
| 20080052123 | Iliff | Feb 2008 | A1 |
| 20080052130 | Iliff | Feb 2008 | A1 |
| 20080052132 | Iliff | Feb 2008 | A1 |
| 20080052318 | Iliff | Feb 2008 | A1 |
| 20080059227 | Clapp | Mar 2008 | A1 |
| 20080059232 | Iliff | Mar 2008 | A1 |
| 20080059247 | Iliff | Mar 2008 | A1 |
| 20080072147 | Brown | Mar 2008 | A1 |
| 20080097180 | Brown | Apr 2008 | A1 |
| 20080097181 | Brown | Apr 2008 | A1 |
| 20080103377 | Brown | May 2008 | A1 |
| 20080108888 | Brown | May 2008 | A1 |
| 20080109172 | Brown | May 2008 | A1 |
| 20080162393 | Iliff | Jul 2008 | A1 |
| 20080262557 | Brown | Oct 2008 | A1 |
| 20080269571 | Brown | Oct 2008 | A1 |
| 20090007924 | Iliff | Jan 2009 | A1 |
| Number | Date | Country |
|---|---|---|
| 1291749 | Apr 2001 | CN |
| 1477581 | Feb 2004 | CN |
| 4430184 | Mar 1995 | DE |
| 4430164 | Feb 1996 | DE |
| 0 320 749 | Dec 1988 | EP |
| 0 447 710 | Sep 1991 | EP |
| 0 531 889 | Mar 1993 | EP |
| 0 643 360 | Mar 1996 | EP |
| 0 720 336 | Jul 1996 | EP |
| 03191952 | Aug 1991 | JP |
| 3202047 | Sep 1991 | JP |
| 04015035 | Jan 1992 | JP |
| 04015035 | Jan 1992 | JP |
| 4056561 | Feb 1992 | JP |
| 06083847 | Mar 1994 | JP |
| 06274472 | Sep 1994 | JP |
| 08117210 | May 1996 | JP |
| 08140944 | Jun 1996 | JP |
| 08164127 | Jun 1996 | JP |
| 08275927 | Oct 1996 | JP |
| WO 9323819 | Nov 1993 | WO |
| WO 9400817 | Jan 1994 | WO |
| WO 9406088 | Mar 1994 | WO |
| WO 9506296 | Mar 1995 | WO |
| WO 9506298 | Mar 1995 | WO |
| WO 9519604 | Jul 1995 | WO |
| WO 9622577 | Jul 1996 | WO |
| WO 9705553 | Feb 1997 | WO |
| WO 9802836 | Jan 1998 | WO |
| WO 9840835 | Sep 1998 | WO |
| WO 9952025 | Oct 1999 | WO |
| WO 0032088 | Jun 2000 | WO |
| WO 0161616 | Aug 2001 | WO |
| WO 0185021 | Nov 2001 | WO |
| WO 0239250 | May 2002 | WO |
| WO 0242876 | May 2002 | WO |
| WO 03040879 | May 2003 | WO |
| WO 03040964 | May 2003 | WO |
| WO 03040965 | May 2003 | WO |
| WO 03040989 | May 2003 | WO |
| WO 03040990 | May 2003 | WO |
| Entry |
|---|
| Alexander G., Health Risk Appraisal, Intern Electro J Health Edu., 2000, 3(Special): 122-137. |
| Ellis et al., Health Education Using Microcomputers II: One year in the Clinic; Preventive Medicine, 1982, 11: 212-224. |
| Ellis et al., Health Education using Microcomputers: Initial Acceptability, Preventive Medicine, Jan. 1981, 10(1): 77-84. |
| Fielding J., Appraising the Health of Health Risk Appraisal, Am J Pub Health, Apr. 1982, 72(4): 337-340. |
| Goetz et al., Health Risk Appraisal: The Estimation of Risk, Health Promotion at the Worksite, Mar.-Apr. 1980, 95(2): 119-126. |
| Replacement Statement and Explanation dated Nov. 21, 2009 from Request for Ex Parte Reexamination of U.S. Patent No. 6,116,540; U.S. Appl. No. 90/009594, filed Nov. 21, 2009. |
| Gale et al., Medical Diagnosis From Student to Clinician, p. 1-22 (1983). |
| Gini et al., “A Serial Model for Computer Assisted Medical Diagnosis,” Int. J. Bio-Medical Computing (11) (1980) pp. 99-113. |
| Office Action for Japanese Patent Application No. 506146/98 (and English language translation). |
| Applied Medical Informatics, Inc., “Medical House CallTM Interactive Home Medical Guide & Symptom Analysis”, Applied Medical Informatics, 1995, pp. 24-Jan., Salt Lake City, UT. |
| Barnett et al., A computer-based medical information system for ambulatory care, Proc. IEEE, 1979, Issue 67, pp. 1226-1237. |
| Bouhaddou et al., An interactive patient information and education system (Medical HouseCall) based on a physician expert system (Iliad), Medinfo, 1995, vol. Pt 2, Issue 8, pp. 1181-1185, Vancouver, Canada. |
| Bouhaddou et al., AMIA, Inc., pp. 742-746, 1995, “Iliad and Medical House Call: evaluating the impact of common sense knowledge on the diagnostic accuracy of a medical expert system”. |
| Collen, Machine diagnosis from a multiphasic screening program, Proceedings of 5th IBM Medical Symposium at 131, 1963. |
| Cope, for well-connected in study, computer's diagnosis is just a phone call away, Minneappolis Star-Tribune, Mar. 1992, pp. 03E. |
| Crossman, Confused? Take two aspirin and call up advisor, New Jersey Record, Apr. 1992, pp. B02. |
| Dawson, Sun, Microsoft battle over Net computing, Multichannel News, Nov. 1996. |
| Gorry et al., Decision analysis as the basis for computer-aided management of acute renal failure, Am. J. Med., Oct. 1973, vol. 3, Issue 55, pp. 473-484. |
| Lai, Abstraction models at system level for interactive multimedia scripting, Master's Thesis, Massachusetts Institute of Technology, May 1995, Boston. |
| Magnet, Who's winning the information revolution, Fortune, Nov. 30, 1992, vol. 12, Issue 126, pp. 110-117. |
| Mallya et al., Correlation in rheumatoid arthritis of concentrations of plasma C3d, serum rheumatoid factor, immune complexes and C-reactive protein with each other and with clinical features of disease activity., Clin. Exp. Immunol., 1982, Issue 48, pp. 747-753. |
| Markoff, Making the PC come alive, New York Times, Sep. 1995. |
| Miller, Dial 1-900 for doctor, Newsweek, Oct. 1991. |
| Okada, Medical data base system with an ability of automated diagnosis, Computer Programs in Biomedicine, Sep. 1977, vol. 3, Issue 7, pp. 163-170. |
| Olson et al., 21st century learning and health care in the home: creating a national telecommunications network, IAF/CRI, Jan. 1992. |
| Pauker et al., Towards the simulation of clinical cognition: taking a present illness by computer, Am. J. Med., Jun. 1976, vol. 7, Issue 60, pp. 981-996. |
| Riordan, Patents; Prodigy's patent is being debated as a possible threat to Sun Microsystems' Java language, New York Times, Feb. 1996. |
| Roberts, Dr. Schueler's home medical advisor 2.0, Compute!, Oct. 1992, Issue 145, pp. 106. |
| Shannon, Peripherals; choosing a college, New York Times, Jan. 1989. |
| Shannon, “Peripherals; advice on a disk: the doctor is really in”, New York Times, Jul. 14, 1992 at C7. |
| Shannon, “Peripherals; Doctor, I have this funny pain . . . ”, New York Times, Nov. 3, 1992. |
| Stearn et al., A statistical analysis of subjective and objective methods of evaluating fabric handle Part 2: Relationship between subjective and objective measurements, Journal of the Textile Machinery Society of Japan, 1988, vol. 2, Issue 34, pp. 39-46. |
| Szolovits et al., Categorical and probabilistic reasoning in medical diagnosis, Artificial Intelligence, Aug. 1978, vol. 1&2, Issue 11, pp. 115-144. |
| Tedesco, Microsoft, Intel and Sun advance NC visions, Broadcasting & Cable, Nov. 1996. |
| Templeton, Medical software that makes house calls, Business Week, Jun. 1992, Issue 3720. |
| Vaughn et al., Effective algorithm-based triage and self-care protocols: quality medicine at lower costs, Ann. Emerg. Med., Jan. 1980, vol. 1, Issue 9, pp. 31-36. |
| Warner, Knowledge sectors for logical processing of patient data in the help system, Proc. IEEE, 1978. |
| Wijkstra et al., Relation of lung function, maximal inspiratory pressure, dyspnoea, and quality of life with exercise capacity in patients with chronic obstructive pulmonary disease, Thorax, May 1994, vol. 5, Issue 49, pp. 468-472. |
| Zallen, Member-centered managed care and the new media, ed. Linda M. Harris, in Health and the New Media, 1995. |
| Gardner, Integrated Computer Systems for Monitoring of the Critically III, Proceedings of 1st Annual Symposium on Computer Application in Medical Care, Washington, D.C., IEEE Computer Society, 1977, pp. 301-302. |
| Koska, Mary T., Primary Care: Hospitals Begin to Target Community Needs, Hospitals, Apr. 5, 1990, 64(7): 24-28. |
| The Alpha Media Catalog, Advertisement, Oct. 1993, “Physician's Database Manager” and “Iliad.” |
| Arthur D. Little, Inc., Acorn Park, Cambridge, MA, Jul. 1992, “Can Telecommunications Help Solve America's Health Care Problems?” Summary. |
| Arthur D. Little, Inc., Acorn Park, Cambridge, MA, Jul. 1992, “Telecommunications: Can It Help Solve America's Health Care Problems?” pp. 1-116. |
| Becher, Ernst, “Fernmeldewesen für soziale Dienste in Entwicklungsländern,” NTZ, 33:304, 1980. |
| Belzer et al., “Encyclopedia of Computer Science and Technology”, Marcel Dekker, Inc., NY (US), 1978, pp. 78-79 and 114-115. |
| Bergman, “Computers make ‘house calls’ to patients; Harvard Community Health Plan offers computerized information service to patients,” J American Hospital Association, 67(10): 52, May 20, 1993. |
| Bowden, K.F. et al., Information Processing, 71:1398-1406, 1972, “Data structures for general practice records.” |
| Cimino, James J. et al., IEEE/Ninth Annual Conference of the Engineering in Medicine and Biology Society, 1987, “DXplain: An interactive knowledge base for assistance in medical diagnostic decisions.” |
| Conrath, David W. et al., IEEE Transactions on Communications, COM-23(10):1119-1126, 1975, “A preliminary evaluation of alternative telecommunication systems for the delivery of primary health care to remote areas.” |
| Currid, Cheryl, PC Week, Sep. 17, 1990, “Risky Business: doctors, lawyers shy away from computer technology.” |
| Dahmer J., “Anamnese und Befund”, Georg Thieme Verlag, 1981, pp. 22-25, 44-47, 292-295 and 371-375 (with English language translation of pp. 22-24, 44-45, 292, 295). |
| Dahmer J., Der Denkprozess “Vom Symptom zur Diagnose”, Anamnese und Befund, Georg Thieme Verlag, 1981, p. 3 (with English language translation) pp. 3. |
| Doheny, Kathleen, La Times Magazine, p. 8, Aug. 4, 1991, “Hack attack.” |
| Fisher et al., “Great Expectations: Expectation-based reasoning in Medical Diagnosis,” Proceedings of the 12th Annual Symposium on Computer Applications in Medical Care, Nov. 6, 1988, pp. 38-42. |
| Frenger, Paul, IEEE Frontiers of Computers in Medicine, 1982, “Details of a database management system for a telephone medical consultation service.” |
| Frenger, Paul, ISA, pp. 103-107, 1983, “Advanced techniques used to create a telephone medical consultation service.” |
| Freudenheim, Milt, The New York Times, Business and Health, p. D2, Jun. 25, 1991, “Computer says take 2 aspirin.” |
| Goldbard, Gary A, Class Notes, Tulane Medicine, Tulane University Medical Center, 1430 Tulane Avenue, New Orleans, LA 70112-2699, p. 26, Jun. 1991. |
| Gome, Amanda, Herald-Sun, p. 13, Nov. 19, 1991, “A picture of success.” |
| Gorry, G. Anthony, Bulletin of the Operations Research Society of America, 19(2), 1971, “FA6.3 Automating Judgmental decision making in medicine.” |
| “Harvard Community Health Plan Testing Computerized Service that Answers Health-Care Questions,” Technical Computing, 6(9), Aug. 1991. |
| Hile, et al. “Reliability of an Automated Decision Support System for Behavioral Treatment Planning: Preliminary Results from the Mental Retardation-Expert”, Computers in Human Services, 10(4): 19-29, 1994. |
| Hudson et al., “Human-Computer Interaction in a Medical Decision Support System.” IEEE Computer Society Press, 2: 429-435, 1989. |
| Kerr, Jennifer, San Diego Union-Tribune, p. A3, Sunday, Jul. 18, 1993, “Phone is link to health-care information.” |
| Larsson et al., “An Expert System Interface for an Identification Program,” Automatica, Pergamon Press Ltd., Oxford, GB., 27(6): 919-930, 1991. |
| Laughlin, Michael L., ed., Computers in Health Care, pp. 32-37, Nov. 1992, “Telecommunications may offer poor a ‘road’ to healthcare.” |
| Levin, Carol, PC Magazine, p. 32, Mar. 16, 1993, “Patient, heal thyself”. |
| Mallory, Jim, Newsbytes, American Association for the Advancement of Science, Panel Discussion, Feb. 19, 1992, “Computers now giving medical advice.” |
| McDonald et al., Environmental Science and Policy Institute, 1992, “Health in the Information Age: The Emergence of Health Oriented Telecommunication Applications.” |
| “Netscape & Sun Announce Javascript The Open, Cross-Platform Object Scripting Language for Enterprise Networks and the Internet”, Press Release, Dec. 4, 1995, web at http://java.sun.com/pr/1995/12/pr951204-03.html. (10 pages). |
| New York Times, p. 18, Jul. 13, 1991, “System helps doctors keep up to date.” |
| O'Neil, et al., Conference Paper, IEEE Coll. on Computer Based Diagnosis, p. 8/1-4, 1989, “Diagnostic Support in the Oxford System of Medicine.” |
| Rose, J, ed., “Progress of Cybernetics, vol. 2, Cybernetics and Industry, Social and Economic Consequences, Cybernetics and Artifacts,” Proceedings of the First International Congress of Cybernetics, London, Gordon and Breach Science Publishers, pp. 803-811, 1969. |
| Rymon, et al., IEEE Transactions on Systems, Man, and Cybernetics, 23(6):1551-1560, Nov./Dec. 1993, “Progressive Horizon Planning-Planning Exploratory-Corrective Behavior.” |
| Sacks, Terry, San Diego Union-Tribune, p. E-16, Mar. 24, 1992, “Pocket computer may cure technology-shy physicians.” |
| Salvans, P. Ferrer and Alonso L. Vallès, Computer Biol. Med., 20(6):433-443, 1990, “An epidemiologic approach to computerized medical diagnosis—AEDMI program.” |
| San Diego Emergency Physicians Society, Meeting Minutes, Regular Oct. 1991 Meeting, P.O. Box 16685, San Diego, CA 92176, first page. |
| Schild, W. et al., IBM J. Res. Develop., 22(5):518-532, 1978, “Computer-aided diagnosis with an application to endocrinology.” |
| Shapiro, Encyclopedia of Artificial Intelligence, 2nd Edition, vol. 2, pp. 916-926, John Wiley & Sons, Inc., 1992. |
| Shortliffe, Edward H., Expert Systems and AI Applications, pp. 323-333, 1980, “Consultation system for physicians: the role of artificial intelligence techniques.” |
| Sloane, L., New York Times, p. 16, Jul. 13, 1991, “For round-the-clock diagnosis, just pick up your telephone.” |
| Smothers, R., New York Times, Sep. 16, 1992, “New video technology lets doctors examine patients many miles away.” |
| Starr et al., “Gycon: A Microcomputer Based Gynecological Consultant,” Proceed Inter Comp Symposium, Dec. 17-19, 1986, Tainan, Taiwan R.O.C. pp. 1678-1684. |
| Starr et al., “A microcomputer-based medical expert system shell using a weight/threshold decision mechanism,” Proc. of the 13th Annual Northeast Bioengineering Conference, Mar. 12-13, 1987, pp. 279-281. |
| Szolovits et al., “Artificial Intelligence in Medical Diagnosis,” Ann Intern Med., Jan. 1988, 108(1): 80-87; (pp. 1-12). |
| Thorpe, C. William et al., “A microcomputer-based interactive cough sound analysis system”, Computer Methods and Programs in Biomedicine, Section II, Systems and programs, 36:33-43, 1991. |
| Wagner, J et al., Conference Paper of Expert Systems and Decision Support in Medicine, 33rd Annual Meeting of the GMDS EFMI Special Topic Meeting, pp. 449-465, Sep. 1988, “A knowledge-based system for interactive medical diagnosis encoding.” |
| Walz, Nancy, The Associated Press, Business News, Jun. 25, 1991, “Computer system aims to wipe out medical paperwork.” |
| Waterman, A Guide to Expert Systems, Addison-Wesley Publishing Co., pp. 46-47 and 272-288, 1986. |
| Weinstock, Edward, Cover, Avant-Garde, 1984, “An Apple a Day™.” |
| Werner, et al., Conference Paper, IEEE Engineering in Medicine and Biology, 3 pages, 1989, “Interlocutor: Conferring with an Expert Diagnostic Consultant in Geriatric Psychiatry .” |
| EPO Examination Report dated Apr. 11, 2005 in European Patent Application No. 02075042.8, filed Jan. 7, 2002. |
| International Search Report and Written Opinion dated Feb. 21, 2007 in PCT/US06/24090, filed Jun. 20, 2006. |
| International Preliminary Report on Patentability dated Jul. 14, 2008 in PCT/US06/24090, filed Jun. 20, 2006. |
| Curtin et al. “Disease Management Information System: Design, Development, Testing, and Clinical Application for Cancer Management”, Abstract http://ascobeta.infostreet.com/prof/me/html/abstracts/hre/m—1509.htm (1997). |
| Medical Computer Consultants' Consortium, Inc. “Disease State Management Software System DMS2—Product Description”, http://www.mc3co.com/DMSS.htm (1997). |
| Memorial Sloan-Kettering Cancer Center “Center Develops New System for Disease Management” http://www.mskcc.org/document/cn950601.htm (1997). |
| Ball et al. Eds. Computers in Health Care: Aspects of the Computer-based Patient Record, Springer Verlag, 1992, pp. 1-336. |
| Barr et al., (Eds.) The Handbook of Artificial Intelligence, HeurisTech Press, 1982, vol. II, Chapter VIII, pp. 175-222. |
| Bischoff, A Knowledge Based System for Assisting in Differential Diagnosis of Chemically Dependent/Mentally Ill Patients, Computers in Human Services, 1992, vol. 8, Nos. 3/4, pp. 143-151. |
| Bortolan et al., The role of patient history in a decision support system, IEEE, Computers in Cardiology, Sep. 1990, Proceedings, pp. 357-360. |
| Brown et al., Information Infrastructure Task Force, The National Information Infrastructure: Agenda for Action, Sep. 15, 1993, pp. 26. |
| Christine, The Future of Health Care Technologies, HighBeam Research, Risk Management, Nov. 1, 1992, 1-3. |
| Common Carrier Week 8, Home Education and Health Benefits said to be underestimated, Warren Publishing Inc., Jan. 20, 1992, 9(3): 1-3. |
| Creative Strategies International (CSI), The Emerging Self-Help Healthcare Market—Microcomputer Applications, 1984, pp. 130. |
| Evans, C. Edward, A Computer in the Waiting Room: Who Needs the Doctor?, Can Fam Phys., Apr. 1984, 30: 869-876. |
| Fallon et al., A Primer for Writing Medical Data Base for the Clinical Decision Support System, Computers and Brains, Progress in Brain Research, Eds. Schadé et al., vol. 33, pp. 155-175. |
| Federal Register, National Telecommunications and Information Administration, DOC—Administration Policy Statement, Notices, Sep. 21, 1993, Fed. Reg., 58(181): 49025-49036. |
| First Opinion Corporation, Canadian Trademark Application File History TMA447,669, registered Sep. 15, 1995 for the Trademark/Servicemark: First Opinion, pp. 51. |
| Haug et al., Decision Support in Medicine: Examples from the HELP System, Computers and Biomedical Research, 1994, 27: 396-418. |
| Henderson, A Trainable Pattern Classifier for Medical Questionnaires, Annals of Biomed Engin., Jan. 27, 1972, 1: 115-133. |
| Johnson et al., Psychological Systems Questionnaire: An Objective Personality Test designed for on-line computer Presentation, Scoring, and Interpretation, Behav Res Meth Instrument., 1979, 11(2): 257-260. |
| Lunin, Lois F., On Speaking Terms With the Computer, Information Today, Feb. 1992, 9(2): 19-20. |
| Malcolm et al., Computer-Assisted Diagnosis of Alcoholism, Computers in Human Services, 1989, 5(3/4): 163-170. |
| McNish, David A., EMED Electronic Medical Database, DMC Software Packaging and Manual, DMC Software Company., 1987, 7 pages. |
| Miller et al., The Computerized Carroll Rating Scale, Indiana University School of Medicine, 1985, pp. 344-347. |
| Partin, A Preliminary Conceptual Framework for the Design, Development, and Use of Client-Oriented Information Systems in Health, J Med Sys. 1987, 11(2/3): 205-217. |
| Patel et al., A Computer-Based, Automated, Telephonic System to Monitor Patient Progress in the Home Setting, J Med Sys., 1992, 16(2/3): 101-112. |
| Shneiderman, Ben, Touch Screens now offer Compelling Uses, IEEE Software, 1991, pp. 93-94. |
| Shortliffe et al., A Model of Inexact Reasoning in Medicine, 1975, (shortened/editied version) Mathematical Biosciences, 11: 233-262. |
| Stead et al., Computer-Assisted Interview of Patients with Functional Headache, Arch Intern Med., 1972, 129: 950-955. |
| Szolovits et al., Guardian Angel: Patient-Centered Health Information Systems, May 1994, MIT Laboratory for Computer Science, TR-604, pp. 40. |
| The National Information Infrastructure: Agenda for Action, U.S. Government Paper, 1993, pp. 42. |
| Walmsley et al., Normal “Anion Gap” (Hyperchloremic) Acidosis, Case Reports, Clin Chem., 1985, 31(2): 309-313. |
| Weinstock, Edward, an Apple a Day . . . ™, Computer Program Manual, Avant-Garde Publishing Corporation, 1984, pp. 47. |
| Weiss et al., Glaucoma Consultation by Computer, Comput Biol Med . . . , Pergamon Press, 1978, 8: 2540. |
| Request for Ex Parte Reexamination filed Oct. 8, 2009 of USP 6,113,540, issued Sep. 5, 2000. |
| Number | Date | Country | |
|---|---|---|---|
| 20060135859 A1 | Jun 2006 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60621384 | Oct 2004 | US |
