The present disclosure relates to an apparatus and method for generating a care plan for a subject.
In post-hospital (post-discharge) management systems currently available, physicians can create a personalized care plan for each patient who joins the post-discharge management program. For each patient, a care plan may include a number of separate sub-plans. For example, for atypical coronary artery disease (CAD) patient, a care plan may include a medication plan, a home monitoring plan, an education and coaching plan, and a follow-up plan, etc. It is time-saving for the physician if they can quickly reach an ideal or intended care plan from a recommended one provided by an automated system, while being able to view the reasoning of the recommendation for reference purposes.
In some currently available solutions, pre-defined protocols, such as possible care elements or possible attributes of care elements in a care plan, can be shown in a list and can be selected by physicians to form an initial plan template. However, in many cases, for certain types of patients the protocols cannot be pre-defined and in some cases the pre-defined protocols cannot cover all possible care plans for a large number of subjects.
Moreover, in some currently available solutions, users or medical personnel (e.g. physicians) are allowed to save existing care plans individually as templates in a database. The saved templates can be searched by users and selected as an initial care plan template for a new patient. However, in this case, there are too many templates in the database which may make it difficult for a user to select a suitable template.
As noted above, there are a number of disadvantages associated with the currently available post-hospital management systems for selecting a care plan for a subject. For example, the management systems that employ pre-defined protocols cannot fully cover all possible care plans for a large number of subjects. As another example, the users of management systems that use historical care plans as templates may find it difficult to navigate through the plurality of care plan templates for a match with a new subject. Moreover, it is difficult and time-consuming for users to manually set up a detailed executable care plan using these management systems, especially when the parameters, i.e. attributes of the care elements in the care plans, available for selection are not as close as possible to the final ideal care plan so a user may have to either select a less accurate parameter or not select a parameter at all.
It would therefore be advantageous to provide an improved method for generating a care plan for a subject.
To better address one or more of the concerns mentioned earlier, in a first aspect, a computer-implemented method for generating a care plan for a subject is provided. The method comprises accessing a plurality of care plans stored in one or more databases, wherein the plurality of care plans are grouped into a plurality of care plan categories based on a similarity measure between one or more attributes of the plurality of care plans, and wherein each of the plurality of care plans is associated with a reference subject. The method also comprises identifying a care plan category for the subject from the plurality of care plan categories using a trained classifier and generating a care plan for the subject based on the one or more care plans stored in the one or more databases that are grouped into the identified care plan category, by maximizing a similarity measure between one or more attributes of the subject and one or more corresponding attributes of the reference subjects.
In some embodiments, the similarity measure between one or more attributes of the plurality of care plans may be based on a number of attributes the one or more care plans have in common and/or that differ by less than a threshold. In these embodiments, two care plans with a similarity measure exceeding a predetermined minimum similarity threshold may be grouped in the same care plan category. In some embodiments, the similarity measure between the one or more attributes of the subject and the one or more corresponding attributes of the reference subjects may be based on the number of attributes the subject has in common with the reference subject and/or that differ by less than a threshold.
In some embodiments, at predetermined time intervals, the plurality of care plans may be regrouped into an updated plurality of care plan categories and the classifier may be retrained.
In some embodiments, the method may further comprise outputting the identified care plan category and the one or more care plans that are grouped into the identified care plan category. In some embodiments, a plurality of care plans may be grouped into the identified care plan category, and the method may further comprise outputting the similarity measures between the care plans that are grouped into the identified care plan category.
In some embodiments, each reference subject may be labelled with a label that identifies the care plan category comprising the care plan for the reference subject. In these embodiments, the method may further comprise: filtering reference subjects from the one or more databases based on the identified care plan category to select candidate reference subjects. The candidate reference subjects may comprise reference subjects with a label identifying the reference subjects as having care plans grouped into the identified care plan category. Moreover, in these embodiments, generating a care plan for the subject may comprise: generating a care plan for the subject, based on the one or more care plans stored in the one or more databases that are grouped into the identified care plan category, by maximizing a similarity measure between one or more attributes of the subject and one or more corresponding attributes of the candidate reference subjects.
In some embodiments, the method may further comprise grouping the care plans associated with the candidate reference subjects into a plurality of care plan sub-categories based on a similarity measure between the care plans associated with the candidate reference subjects. In these embodiments, the method may further comprise training a further classifier to identify a care plan sub-category for the subject from the plurality of care plan sub-categories.
In some embodiments, the method may further comprise compiling an ordered list of the care plans of the identified care plan category, wherein the list is ordered based on a similarity measure between one or more attributes of the subject and one or more corresponding attributes of each of the reference subjects. In these embodiments, the higher the similarity measure between the one or more attributes of the subject and one or more attributes of a reference subject, the higher the care plan associated with the reference subject may be in the list.
In a second aspect, there is provided a computer program product comprising a computer readable medium, the computer readable medium having computer readable code embodied therein, the computer readable code being configured such that, on execution by a suitable computer or processor, the computer or processor is caused to perform the method described above.
In a third aspect, there is provided an apparatus for generating a care plan for a subject. The apparatus comprises a processor configured to access a plurality of care plans stored in one or more databases, wherein the plurality of care plans are grouped into a plurality of care plan categories based on a similarity measure between one or more attributes of the plurality of care plans, and wherein each of the plurality of care plans is associated with a reference subject. The processor is further configured to identify a care plan category for the subject from the plurality of care plan categories using a trained classifier and generate a care plan for the subject based on the one or more care plans stored in the one or more databases that are grouped into the identified care plan category, by maximizing a similarity measure between one or more attributes of the subject and one or more corresponding attributes of the reference subjects.
According to the aspects and embodiments described above, the limitations of existing techniques are addressed. In particular, the above-described aspects and embodiments enable an automatic recommendation of a care plan to be performed. The automatic recommendation takes into account similar historical care plans and the similarity between one or more attributes of the subject and one or more corresponding attributes of a number of reference subjects. In this way, prediction errors can be reduced. Moreover, the method does not require the use of predefined protocols or templates. This improves the efficiency and accuracy for care plan generation. There is thus provided an improved method and apparatus for generating a care plan for a subject.
These and other aspects of the disclosure will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.
For a better understanding of the embodiments, and to show more clearly how they may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:
As noted above, there is provided an improved apparatus and a method of operating the same which addresses the existing problems.
For example, a care plan may include a medication plan including a list of medication. In this example, each medication in the list is a care element associated with the “medication” class. Also, in this example, each medication in the list is associated with one or more attributes, e.g. a medicine name, a medicine dose, a medicine frequency, a medicine effect duration period, a time of day for taking the medicine (e.g. morning, evening), a chemical structure of a medicine, a pharmacology of a medicine, a manufacturer of a medicine, etc. A care plan which only comprises care elements with attributes that are highly clinically-relevant is referred to as a “high-level care plan”, and a care plan which comprises care elements with the necessary attributes to be executed by a subject is referred to as a “detailed care plan”.
A care plan may comprise other types of plans (i.e. plans associated with other classes), such as a nutrition care plan, an exercise care plan, a mental health care plan, a home monitoring plan, an education and/or coaching plan, a follow-up plan. The care elements contained in these types of plans are respectively associated with classes such as “nutrition care”, “exercise”, “mental health”, “home monitoring”, “education and/or coaching”, and “follow-up”.
As illustrated in
Briefly, the processor 102 is configured to access a plurality of care plans stored in one or more databases. The plurality of care plans are grouped into a plurality of care plan categories based on a similarity measure between one or more attributes of the plurality of care plans and each of the plurality of care plans is associated with a reference subject. The processor 102 is also configured to identify a care plan category for the subject from the plurality of care plan categories using a trained classifier. The processor 102 is further configured to generate a care plan for the subject based on the one or more care plans stored in the one or more databases that are grouped into the identified care plan category, by maximizing a similarity measure between one or more attributes of the subject and one or more corresponding attributes of the reference subjects.
The one or more attributes of the subject referred to herein may, for example, comprise any one or more of: clinical attributes of the subject, demographic attributes of the subject, and administrative attributes of the subject. For example, the one or more attributes of the subject may comprise a diagnosis of the subject (which is an example of a clinical attribute), an age of the subject (which is an example of a demographic attribute), and/or a physician of the subject (which is an example of an administrative attribute).
In some embodiments, the apparatus 100 may further comprise at least one user interface 104. Alternative or in addition, at least one user interface 104 may be external to (i.e. separate to or remote from) the apparatus 100. For example, at least one user interface 104 may be part of another device. A user interface 104 may be for use in providing a user of the apparatus 100 with information resulting from the method described herein. For example, the processor 102 may be configured to control one or more user interfaces 104 to render (or output or display) the generated care plan(s) for the subject. Alternative or in addition, a user interface 104 may be configured to receive a user input. For example, a user interface 104 may allow a user of the apparatus 100 to manually enter instructions, data, or information. In these embodiments, the processor 102 may be configured to acquire the user input from one or more user interfaces 104.
A user interface 104 may be any user interface that enables the rendering (or output or display) of information to a user of the apparatus 100. Alternatively or in addition, a user interface 104 may be any user interface that enables a user of the apparatus 100 to provide a user input, interact with and/or control the apparatus 100. For example, the user interface 104 may comprise one or more switches, one or more buttons, a keypad, a keyboard, a touch screen or an application (for example, on a tablet or smartphone), a display screen, a graphical user interface (GUI) or other visual rendering component, one or more speakers, one or more microphones or any other audio component, one or more lights, a component for providing tactile feedback (e.g. a vibration function), or any other user interface, or combination of user interfaces.
In some embodiments, the apparatus 100 may comprise a memory 106. Alternatively or in addition, one or more memories 106 may be external to (i.e. separate to or remote from) the apparatus 100. For example, one or more memories 106 may be part of another device. A memory 106 can be configured to store program code that can be executed by the processor 102 to perform the method described herein. A memory can be used to store information, data, signals and measurements acquired or made by the processor 102 of the apparatus 100. For example, a memory 106 may be used to store (for example, in a local file) a plurality of care plans and one or more attributes of the plurality of care plans. The processor 102 may be configured to control a memory 106 to store the plurality of care plans and one or more attributes of the plurality of care plans.
In some embodiments, the apparatus 100 may comprise a communications interface (or circuitry) 108 for enabling the apparatus 100 to communicate with any interfaces, memories and/or devices that are internal or external to the apparatus 100. The communications interface 108 may communicate with any interfaces, memories and/or devices wirelessly or via a wired connection. For example, the communications interface 108 may communicate with one or more user interfaces 104 wirelessly or via a wired connection. Similarly, the communications interface 108 may communicate with the one or more memories 106 wirelessly or via a wired connection.
It will be appreciated that
With reference to
In addition, the plurality of care plans are grouped into a plurality of care plan categories based on a similarity measure between one or more attributes of the plurality of care plans. This grouping may be referred to as “clustering” and may be performed using a clustering algorithm. A person skilled in the art will be aware of various clustering algorithms that may be used to group the plurality of care plans into a plurality of care plan categories. However, in some embodiments, for example, the Tanimoto Coefficient may be used, as part of a clustering algorithm, for the grouping of the plurality of care plans. In some embodiments, the plurality of care plans that are grouped into a plurality of care plan categories may be “high-level care plans”, which only comprise care elements that are highly clinically-relevant.
As mentioned above, a care plan can comprise a plurality of care elements with attributes presented in the same information structure. Each care element may be associated with a specific class and one or more attributes. In some embodiments, the similarity measure between one or more attributes of the plurality of care plans may be based on a number of attributes the one or more care plans have in common and/or that differ by less than a threshold. In some embodiments, at block 202 of
For example, the similarity measure between one care plan and another care plan in a plurality of care plans may be represented by a number indicative of the number of attributes that the two care plans have in common, and the predetermined minimum similarity threshold may be an integer number (such as 5 or any other integer number). Thus, in this example, two care plans are grouped into the same care plan category where the two care plans have at least the integer number of attributes in common. In some embodiments, grouping of the plurality of care plans into a plurality of care plan categories may comprise, for each of the plurality of care plans, determining the similarity measure between the care plan and each of the rest of the plurality of care plans and then comparing the determined similarity measures against the predetermined minimum similarity threshold. In an example where the predetermined minimum similarity threshold is an integer number of 5, if the similarity measure between a first care plan P1 and a second care plan P2 is 6, then the processor 102 groups the first care plan P1 and the second care plan P2 in the same care plan category. In the same example, if a similarity measure between the first care plan P1 and a third care plan P3 is 3, then the processor 102 does not group the first care plan P1 or the second care plan P2 with the third care plan P3 in the same care plan category. Thus, the third care plan P3 in this example is grouped in a different care plan category to the first care plan P1 and the second care plan P2.
Returning back to
Although not illustrated in
Also, although not illustrated in
Returning back to
As an example, if at block 204 of
In some embodiments, the similarity measure between the one or more attributes of the subject and the one or more corresponding attributes of the reference subjects may be based on the number of attributes the subject has in common with the reference subject and/or that differ by less than a threshold. As mentioned above, the one or more attributes of the subject may comprise any one or more of: clinical attributes of the subject (e.g. a diagnosis), demographic attributes of the subject (e.g. an age), and administrative attributes of the subject. Therefore, in an example, if the subject has the same age as a reference subject this may be indicated by an increased value in the similarity measure between the one or more attributes of the subject and the one or more attributes of the reference subject. For example, a similarity measure between the one or more attributes of the subject and one or more corresponding attributes of a reference subject may be represented by a value which indicates the number of attributes the subject has in common with the reference subject, e.g. a value of “4” indicating that the subject has four attributes in common with the reference subject.
The computer-implemented method in some embodiments may further comprise compiling an ordered list of the care plans of the care plan category identified at block 204 of
In some embodiments, each reference subject may be labelled with a label that identifies the care plan category comprising the care plan for the reference subject. In these embodiments, the computer-implemented method may further comprise filtering reference subjects from one or more databases based on the care plan category identified at block 204 of
In some of the embodiments involving candidate reference subjects, the care plans associated with the candidate reference subjects may be grouped into a plurality of care plan sub-categories based on a similarity measure between the care plans associated with the candidate reference subjects. The grouping of the care plans into care plan sub-categories may be based on a similarity measure between one or more attributes of the plurality of care plans. In these embodiments, the computer-implemented method may further comprise training a further classifier to identify a care plan sub-category for the subject from the plurality of care plan sub-categories.
Although not illustrated in
There is thus provided an improved method and apparatus for generating a care plan for a subject, which overcomes the existing problems.
There is also provided a computer program product comprising a computer readable medium, the computer readable medium having computer readable code embodied therein, the computer readable code being configured such that, on execution by a suitable computer or processor, the computer or processor is caused to perform the method or methods described herein. Thus, it will be appreciated that the disclosure also applies to computer programs, particularly computer programs on or in a carrier, adapted to put embodiments into practice. The program may be in the form of a source code, an object code, a code intermediate source and an object code such as in a partially compiled form, or in any other form suitable for use in the implementation of the method according to the embodiments described herein.
It will also be appreciated that such a program may have many different architectural designs. For example, a program code implementing the functionality of the method or system may be sub-divided into one or more sub-routines. Many different ways of distributing the functionality among these sub-routines will be apparent to the skilled person. The sub-routines may be stored together in one executable file to form a self-contained program. Such an executable file may comprise computer-executable instructions, for example, processor instructions and/or interpreter instructions (e.g. Java interpreter instructions). Alternatively, one or more or all of the sub-routines may be stored in at least one external library file and linked with a main program either statically or dynamically, e.g. at run-time. The main program contains at least one call to at least one of the sub-routines. The sub-routines may also comprise function calls to each other.
An embodiment relating to a computer program product comprises computer-executable instructions corresponding to each processing stage of at least one of the methods set forth herein. These instructions may be sub-divided into sub-routines and/or stored in one or more files that may be linked statically or dynamically. Another embodiment relating to a computer program product comprises computer-executable instructions corresponding to each means of at least one of the systems and/or products set forth herein. These instructions may be sub-divided into sub-routines and/or stored in one or more files that may be linked statically or dynamically.
The carrier of a computer program may be any entity or device capable of carrying the program. For example, the carrier may include a data storage, such as a ROM, for example, a CD ROM or a semiconductor ROM, or a magnetic recording medium, for example, a hard disk. Furthermore, the carrier may be a transmissible carrier such as an electric or optical signal, which may be conveyed via electric or optical cable or by radio or other means. When the program is embodied in such a signal, the carrier may be constituted by such a cable or other device or means. Alternatively, the carrier may be an integrated circuit in which the program is embedded, the integrated circuit being adapted to perform, or used in the performance of, the relevant method.
Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims. In the claims, the word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality. A single processor or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.
Number | Date | Country | Kind |
---|---|---|---|
PCT/CN2017/109695 | Nov 2017 | CN | national |
17203595.8 | Nov 2017 | EP | regional |
This application claims the benefit of European Patent Application No. 17203595.8, filed on 24 Nov. 2017 and International Application No. PCT/CN2017/109695, filed 7 Nov. 2017. These applications are hereby incorporated by reference herein.