Patent Application
20100076786
The provision of health care related information over publicly accessible computer systems, such as the Internet, is generally limited to a few kinds of activities. These activities include: access to personal health records; scheduling appointments, registration and bill payment; secure messaging with service providers; access to general health information such as current medications. In other words, most current systems provide general information and basic transactions.
Access to personal health records, scheduling appointments, registration and bill payment typically is provided through password protected accounts accessed over the Internet. Secure messaging with service providers also may include password protected accounts that a user may access over the Internet. After logging in, a user can send and receive electronic mail, text messages, or instant messages to and from service providers.
Access to general health information typically is provided through web sites accessed over the Internet that provide health care information relating to certain types of disease or populations of patients. For example, one may access a cancer survivors' network, which provides discussion groups and a resource library. The information provided typically is generic for all users and is disease-centric.
General information and basic transactions are not specific enough to each patient to fulfill the needs of patients and their caregivers. Personalized health information is provided by maintaining a patient profile over time and matching information from the patient profile to appropriate patient-centric information related to that patient's specific medical condition. The patient profile includes a variety of medical information, including, but not limited to medical history, diagnostic information, prognostic information, survivorship plan, current treatment regimens and possibly genetic information. The patient profile is personalized, in that it contains not only historical information related to the patient and patient's care, but also information that is matched to the patient based on past, current and/or future care. The patient profile is matched, for example, to patient-centric options for care, clinical trials for which the patient is eligible, drug treatment regimens directly related to each patient's specific condition, scheduling of the treatments, survivorship plans and data, information about managing symptoms, nutritional information, rehabilitation recommendations, possible adverse events based on specific treatments to monitor, and personalized education and information to name a few.
Because the patient profile is updated and modified over time, as the patient receives care, experiences and reports symptoms and outcomes of care using electronic interfaces or clinical visits, the personalized health information is updated periodically. Therefore, the patient or caregiver receives up to date personalized health information when the information is accessed. For example, a patient may report symptoms, which in turn can be used to access recommendations for managing those symptoms in the context of the rest of the patient's medical condition, history and treatment regimen. Patient responses to questionnaires, which facilitate the integration of valuable data from multiple health care providers, also may prompt recommendations for lifestyle modifications or different medicines and treatments.
Referring now to
The computer system 10 includes a portal computer system 12, through which multiple patients (or caregivers) request access to patient-specific information. Each patient, or caregiver, may access the portal computer system 12 through, for example, a personal computer system 14 over a publicly accessible computer network 16 such as the Internet. The computer system 10 can be represented as a computer network and typically includes multiple computers that are logically connected through a variety of communication channels, whether wired or wireless, and that communicate using standard computer network communication protocols such as TCP/IP, UDP and/or HTTP.
When accessing the portal computer system 12, the personal computer system 14 transmits a unique identifier 18 assigned to the patient. This unique identifier 18 is used by the portal computer system 12 to cause information about the patient, in particular personalized health information 28, to be retrieved from a database 22. The database 22 stores, for each patient (as indicated by the unique identifier 18 for the patient), personalized health information 28 for the patient, which also may include information such as the patient profile 20 or other helpful information. The portal computer system 12 transmits at least a portion of this personalized health information 28 to the personal computer system 14. In some implementations, the information provided is limited according to the authorization level for the user, e.g., according to whether the user is the patient, the patient's doctor, some other caregiver for the patient, or other authorized person. It should be understood that the database 22 in
To generate the personalized health information 28, a personalized information generation module 24 uses a patient profile 20 and other information 30 from information sources 26. The personalized health information 28 is stored in the database 22 by module 24. Information from a patient profile 20 also may be stored directly in the database 22. The generation of personalized health information 28 will be described in more detail below in connection with
In one example, the personal computer system 14 is a general-purpose computer with a connection to the Internet and includes at least a display and web browser software. The web browser software contacts the portal computer system 12 and receives from the portal computer system 12 a main document (e.g., a domain top level document), which may be in hypertext markup language (HTML), extensible markup language (XML) or other web browser compatible format, through which a user may log into the portal computer system 12 using a username and password. After the personal computer system 14 submits the username and password to the portal computer system 12, the portal computer system 12 provides the personalized health information 28 to the patient through the personal computer system 14. Examples of user interfaces provided after authentication are shown in
The portal computer system 12, in one example, is a general purpose server computer configured to run web server software and interface software enabling it to connect to database 22 and the Internet.
Although
As another example implementation shown in
In another example implementation, the personalized information generation module 24 can generate the personalized health information 28 in response to the patient accessing the portal computer system 12, in which case the module 24 would receive a unique identifier 18 for that patient to enable it to access a patient profile 20.
The patient profile 20 includes, for example, medical information related to medical history, diagnostic information, prognostic information, treatment regiment, and optionally genetic information for each patient. The patient profile 20, for example, may include information about a patient's genetic disposition to disease, demographics, lifestyle choices, and molecular profiles related to specific diseases of the patient. Other examples of medical information that may be included in a patient profile 20 include, but are not limited to: an electronic medical record, including diagnoses, treatments, prognoses, medicines, lab results and outcomes of care; surgical data, including procedures and timelines; survey information including demographics, history, screening and risk assessment; portions of a cancer registry, including follow-up information, diagnosis and vital status; pathology information including typing information; tissue samples and other biospecimen information, including location, diagnosis, preparation and type; clinical trial data, including protocols, studies and participation; operational data, including billing, scheduling and visit information; and microarray data, including gene expression, experiment analysis and sequence verified annotation; prescriptions written and fulfilled; vital signs; and physician's notes.
The patient profile 20 can be enhanced through directed questionnaires and symptom reports completed by each patient, or a caregiver. For example, a patient or caregiver may report symptoms using a form presented to the user via the personal computer system 14. Various forms can be presented as part of a survivorship plan at times in accordance with a patient's condition/treatment in order to augment the patient profile 20. As another example, in some instances, the generation of personalized health information 28 may require information from the patient that is not in the patient profile 20. A patient or caregiver may be prompted to complete a questionnaire, using a form presented to the user via the personal computer system 14. When such a form is completed it is submitted to the portal computer system 12, which in turn provides data to an appropriate clinician or clinical data system for storage. The information from the form can be included in the patient profile 20.
Other user interfaces are possible.
Returning now to
The datamart 102 can be created by combining patient data from the variety of sources that have patient data, because there are generally multiple sources of patient data for each patient. This combination can be performed using a centralized repository 104 that includes a staging system 106, which periodically copies, cleanses, aggregates and transforms data from information sources 26 into the staging system 106. A global repository 110 periodically is updated with correctly formatted and quality-verified data from the staging system 106. The other sources 108 of data may include, but are not limited to, cancer patient registries of institutions or agencies, tissue sample tracking database, clinical trial databases (such as those provided by the caBIG® (Cancer Biomedical Informatics Grid) network at the National Institutes of Health), electronic medical record databases, financial record (billing and payment) databases, scheduling databases, and other health information systems such as those provided by Galvanon, Inc., Cerner Corporation and others. Some of these sources can be of the form of clinical information systems for which clinical data is entered by clinicians and/or other staff of a health care provider. Data also may come from resources (e.g., the Internet) and from patients. In some implementations, any patient-provided data is not supplied to the online database 100 but instead is placed first into one or more of the clinical data systems (part of information sources 26) that provide data to the central repository 104. In some implementations, the patient profile 20 is generated substantially from information sources and not from the patient directly. The global repository 110 provides a stable representation of the combined patient data. In some implementations, the patient and patient related data is extracted from global repository 110 into datamart 102 on periodic cycles that make the patient data available to other computing resources (e.g., the personalized information generation module 24 or portal computer system 12). As described above, the patient data is accessed through portal computer system 12 when proper authentication is provided by the personal computer system 14. Other datamart(s) 112 also may periodically extract data from the global repository 110.
Having now described the architecture of an example computer system for providing personalized data to a patient, caregiver or other authorized entity, the generation of personalized health information will now be described. Generation of personalized data includes two parts, matching and retrieving information. Referring now to
At stage 706 a patient profile is identified. The patient profile can be of the form of a record that is stored in database 22. In some implementations, the patient profile 20 is a record stored in the global repository 110. At stage 708, patient centric information is identified. In some implementations, the patient centric information has been previously stored in or associated with the patient profile 20. In some implementations, the patient centric information is developed in real time based on other information in the patient profile 20 (i.e., the personalized information generation module 24 can generate in real time information for use in populating one or more pages to be returned to the user in response to the login operation). Whether in real time or cached, information in the patient profile 20 is matched to appropriate patient-centric information related to that patient's specific medical condition to provide the personalized health information 28. The personalized health information 28 may include options for care, clinical trials for which the patient is eligible, drug treatment regimens directly related to each patient's specific condition, scheduling of the treatments, survivorship plans and data, information about managing symptoms, nutritional information, rehabilitation recommendations, possible adverse events based on specific treatments, and personalized education and information, all related to that patient's specific medical condition. This matching of information from a patient profile to provide personalized health information 28 may be implemented in many ways, depending on the type of personalized health information 28.
As one example, a decision tree may be built for each type of personalized health information 28 that may be provided. Each node in the decision tree can be based on information in a patient profile. If information required by a decision tree is not available in the patient profile, the patient or caregiver may be prompted for the information. A decision tree can be represented in computer software and be applied to input data (in this example, medical information from a patient profile) when that computer software is executed by a computer to provide the outputs of the decision tree (in this case, the relevant personalized health information).
Instead of a decision tree, other computational models that map multiple inputs to personalized health information 28 may be used, such as neural networks, Markov processes, support vector machines, or other probabilistic models. The use of decision trees in some cases may be preferable because the reasoning behind results of a decision tree may be more easily explained than the results provided by probabilistic models.
An appropriate decision tree, or other computational model, may be built by analyzing actual patient data using a form of clustering analysis by a computer, by using expert opinion, by using results of reported studies, or any combination of these. An example of clustering analysis includes, but is not limited to, classification and regression trees. Clustering analysis can be performed using data in the database 22 (or any of the data warehouses shown in
In some implementations, by including, in the clustering analysis, information about outcomes of treatment, comorbidity data and survivorship data, a group of similar patients may be further classified by their treatment regimens, by distinguishing treatments that worked from treatments that did not work. A patient that is otherwise similar to one group of patients may be recommended for treatment that has been identified to work for similar patients. Alternatively, if a patient that is similar to one group of patients and is receiving treatment that has been identified as not working for that group of patients, then an alternate treatment regimen can be identified and provided. Outcome data also can be used to provide survival statistics and prognosis for a group of patients, which in turn can be provided as part of the personalized health information 28 of each patient that matches the characteristics of that group.
There are many kinds of personalized health information 28 that can be provided in this manner from the medical information for a patient.
As one example, the personalized health information 28 can include options for care given a patient's current condition and treatment regimen. In this case a computational model of the personalized information generation module 24 maps information from the patient profile 20 to the various options for care available for this patient. If the options for care require an appointment to visit a doctor, the personalized health information 28 may include the ability to schedule appropriate appointments.
As another example, the availability of clinical trials in which the patient can participate also can be part of the personalized health information 28. In this case the computational model of the personalized information generation module 24 can map data in the patient profile 20 to the inclusion and exclusion criteria of available clinical trials. If information for some of the criteria is not available, the personalized health information 28 for the patient may be augmented to include a prompt or other means for retrieving the missing information from either the patient or other caregiver (i.e., a prompt to link to a questionnaire can be provided with the personalized health information 28 that is provided through the portal). The missing information could require, for example, a laboratory test, and the personalized health information 28 for the patient could include a recommendation to the patient that the laboratory test be performed. The ability to schedule the laboratory test also can be provided. Such a system may be implemented in part using the system described in U.S. Patent Publication No. 2008/0033658, which is hereby incorporated by reference.
As another example, the personalized health information 28 can include drug treatment regimens related to a patient's current condition. In this case the computational model of the personalized information generation module 24 maps information from the patient profile 20 to the various drug treatments available for this patient. If the available drug treatments could be scheduled, or require an appointment to visit a doctor for a consultation, the personalized health information 28 may include the ability to schedule appropriate appointments.
As another example, the personalized health information 28 can include scheduling information for treatments already approved for the patient. In this case the computational model of the personalized information generation module 24 maps information from the patient profile 20, such as the approved treatment, to scheduling information for the facility providing the treatment.
As another example, especially for cancer patients, the personalized health information 28 can include survivorship data and a survivorship plan that is updated automatically and includes personalized data based on the patient's disease, previous treatment and other relevant characteristics. Such other characteristics may include a description of long-term and late effects of treatment, recommended surveillance test for late effects, when they should occur, and where they will be performed. Scheduling of appointments for such tests can be provided. Survivorship data can be accessed based on a patient's disease molecular profile (such as a tumor genetic profile) and corresponding treatment. In this example, the patient profile 20 can include tumor genetic profiles and treatments. The computational model of the personalized information generation module 24 can map tumor genetic profiles and treatment information to survivorship data and plans. Using the computational model, survivorship data and plans from similar patients can be processed to provide appropriate survivorship information to the patient. Because the plan is derived from the patient profile 20, it is a dynamic plan that is continually updated as the patient profile 20 changes and as other information used to derive the plan changes. In this example, such a system may be implemented in part using the system described in U.S. Patent Publication No. 2005/0256745, which is hereby incorporated by reference.
As another example, a patient may report symptoms. In this case, the personalized health information 28 may include recommendations for managing those symptoms (e.g., palliative care recommendations) in the context of the rest of the patient's medical condition, history and treatment regimen. In particular, the computational model of the personalized information generation module 24 can map the symptoms, along with other relevant information from the patient profile, such as current treatment regimen and other information about the patient's medical condition, to recommendations for managing symptoms. If the recommendations include, for example, some additional treatment or medicine, then the personalized health information for the patient could include a recommendation to the patient to pursue such treatment or take such medicine. Scheduling of appropriate treatments or appointments to obtain prescriptions can be provided.
As another example, the personalized health information 28 may include nutritional and/or rehabilitation and/or other lifestyle information or recommendations relevant to the patient's medical condition and treatment. In this example, the computational model for the personalized information generation model 24 can map information from the patient profile 20, such as the treatment regimen, symptoms, medical history, etc., to nutritional and rehabilitation recommendations specific to the patient. Notably, this information might be updated in response to other data provided by or about the patient during the course of treatment (such as through the response to a questionnaire).
As another example, the personalized health information 28 may include warning signs or adverse events to look out for that are relevant to the patient's medical condition and treatment. In this example, the computational model for the personalized information generation model 24 can map information from the patient profile 20, such as the treatment regimen, symptoms, medical history, etc., to known warning signs or adverse events that would indicate a deterioration or other change in the patient's condition that might require medical attention.
As another example, the personalized health information 28 may include information extracted from an electronic library, or links to information in an electronic library, that is most relevant to the patient's profile, given his/her disease, treatment, and other personal characteristics. In this case, the computational model can map information from the patient profile 20 to an electronic library, perhaps by using information from the patient profile 20 to perform a search on the electronic library, to provide links to information that the patient or a caregiver may find relevant. The electronic library may include the Internet, and/or indexed data from the Internet. The information that is highly relevant to a patient can be identified using a variety of concept mapping techniques, such as CMapTools Software from the Institute for Human and Machine Cognition (IHMC) for which documentation is provided at http://cmap.ihmc.us/documentation.
As another example, the computational model comprises concept maps for each patient derived from the patient profile 20 for each patient. The personalized information generation module 24 can search for and aggregate information from multiple information sources using the concept maps. A concept map is a technology developed by the Institute for Human and Machine Cognition. The Cmaps tool is a front end user interface that allows a user to more easily navigate a complex website. Concept maps are graphical tools for organizing and representing knowledge. They include concepts (e.g., enclosed in circles or boxes of some type) and relationships between concepts indicated by a connecting line linking two concepts. Words on the line, referred to as linking words or linking phrases, specify the relationship between the two concepts. In some implementations, a concept is defined as a perceived regularity in events or objects, or records of events or objects, designated by a label. In some implementations, the label is a word, although sometimes symbols such as + or %, or more than one word is used. Propositions are statements about some object or event in the universe, either naturally occurring or constructed. Propositions contain two or more concepts connected using linking words or phrases to form a meaningful statement. In some implementations, propositions are referred to as semantic units, or units of meaning. In some implementations, the concepts presented in a concept map are represented in a hierarchical fashion with the most inclusive, most general concepts at the top of the map and the more specific, less general concepts arranged hierarchically below. The hierarchical structure for a particular domain of knowledge also depends on the context in which that knowledge is being applied or considered. Concept maps are described in more detail on the web at http://cmap.ihmc.us/conceptmap.html. In some implementations, personalized information generation module 24 incorporates machine learning methodologies such that the a concept map (Cmap) interface is capable of learning from the user which paths are used more often and direct the presentation of content based on user preferences.
Returning to
The method described in association with
In some implementations, the method can include one or more of the following additional features or steps. Extracting relevant information can include extracting information from a cancer registry. Extracting relevant information can include periodically evaluating the medical record for changes and extracting relevant information if changes are detected. Generating a medical summary can include periodically updating the medical summary based on receipt of new relevant information. In some implementations, the method includes evaluating the medical profile including mapping the medical profile to one or more decision trees and evaluating each decision tree to determine patient centric information; and providing the patient centric information in the portal to the user. Evaluating a decision tree can include determining if all information required to evaluate the decision tree is available, and if not, prompting the user to provide additional information. Prompting can include providing a questionnaire to the user and updating the patient profile to include information from the questionnaire.
Another example method for presenting patient centric information includes the following steps: extracting relevant information from a medical record for a patient; storing the relevant information in a patient profile separate from the medical record and accessible by a user of a networked device; augmenting the patient profile including generating patient centric information, the patient centric information including information that is not part of the medical record, the augmenting including mapping information from the patient profile to one or more templates and evaluating the templates to determine additional information to provide to the patient that is specific to one or more of their current condition, their current treatment, their past treatment, or their expected survival; storing the patient centric information in association with the patient profile; and presenting at least a portion of the profile and the patient centric information to the user in a portal in response to a request received from the user of the networked device that includes a patient identifier and authenticating information.
Yet another example method for presenting patient centric information includes the following steps: extracting information from a database, the extracted information forming a patient profile, the database maintained by an institution and including cancer registry information, the database being periodically updated with new information for both new and existing patients, the database also periodically being evaluated and used to update a local, regional, or national cancer registry not maintained by the institution; maintaining the patient profile separate from the database; generating a medical summary that includes a current medical condition and current treatment plan for a patient using the patient profile; and presenting the medical summary in a portal to a user of a networked device that has presented an identifier associated with the patient and authentication information.
The method can include one or more of the following additional features or steps. The method can include generating a survivorship plan detailing at least future diagnostic procedures for the patient and presenting the survivorship plan in the portal to the user. Presenting the medical summary can further include generating display data that includes the medical summary. The method can further include augmenting the patient profile including generating patient centric information, the patient centric information including information that is not part of the database, the augmenting including mapping information from the patient profile to one or more templates and evaluating the templates to determine additional information to provide to the patient that is specific to one or more of their current condition, their current treatment, their past treatment, or their expected survival.
Referring now to
Referring now to
The techniques and methods described above can be implemented using a combination of computer hardware and computer software. The techniques and methods can be implemented as a computer program product, i.e., computer program instructions encoded on or in a tangible computer-readable medium, for execution by, or to control the operation of, data processing apparatus, e.g., a programmable processor, a computer, or multiple computers. A computer program can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program can be deployed to be executed on one computer or on multiple computers at one site or distributed across multiple sites and interconnected by a communication network. Each computer may implemented using one or more programmable processors executing a computer program to perform functions described herein by operating on input data and generating output, or using special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit).
Processors suitable for the execution of a computer program include, by way of example, both general- and special-purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from a read-only memory or a random access memory or both. The computer includes a processor for executing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto-optical disks, or optical disks. Information carriers suitable for embodying computer program instructions and data include all forms of non-volatile computer memory, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in special purpose logic circuitry.
A computer also may include a variety of input and output devices. Example input devices include but are not limited to a keyboard and a pointing device, such as a mouse, trackball, touchpad or the like. Example output devices include a display or monitor, audio outputs, and printing.
As shown in
Having described example embodiments, it should be apparent to those skilled in the art that the foregoing is merely illustrative and not limiting, having been presented by way of example only. Numerous modifications and other embodiments are with the scope of ordinary skill in the art and are contemplated as falling with the scope of the invention.
This application claims the benefit of priority from U.S. Provisional Application No. 61/086,649, filed Aug. 6, 2008, which provisional application is incorporated by reference herein in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61086649 | Aug 2008 | US |
