Description:DESCRIPTION
Technical Field
[001] Generally, the invention relates to machine learning (ML). More specifically, the invention relates to the method and system for analyzing patient data from the electronic medical record (EMR) using ML models.
BACKGROUND
[002] Current clinical trials take years to design, analyze, and arrive at a useful outcome and a trial result. Clinicians often require patient and disease specific recommendations customized to their patient population and their area of practice. Current research methods are unable to meet this level of tailored therapy. Adaptive trials that use a Bayesian model are fast. However, these trials are not as per the requirement of clinicians. For example, the trials may not be context specific. The trials are also not incorporated into electronic medical records (EMR) for real-time clinical decision support. Further, the trials (i.e., both adaptive and frequentist models) do not consider historical data sets and historical studies for an informed clinical decision making with the historical context. Disease based registries include large sets of data but there is no data available for adaptive clinical trial platforms that make clinically relevant data available rapidly to care providers.
[003] Various decision support systems already exist on the EMR which are usually rule-based. However, these systems are not connected to up to date society guidelines, disease registries, and clinical trial platforms. Also, the systems are not context specific, and tailor made to the patient's care. Hence, the systems are not customized as per the patient or to the doctor requirements.
[004] Therefore, there is a need to develop a system and a method that combines disease specific information (such as a registry) with an adaptive platform that runs on an EMR and has access to existing clinical decision support systems to better inform a clinician.
SUMMARY OF INVENTION
[005] In one embodiment, a method for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models is disclosed. The method may include receiving real-time patient data corresponding to the patient. It should be noted that the real-time patient data may include geographical location of the patient, patient details, comorbidities, details of current medical illness, medications, laboratory data, clinical notes, vitals signs, radiology data and information of a provider associated with the patient. The method may further include associating a built in adaptive trial platform (with novel and/or prevalent statistical tools such as frequentist models or bayesian methodologies) and a disease-specific registry of the patient (and other patients with the same disease profile) with the EMR to monitor the real-time patient data. The method may further include upon associating, analyzing, through a first ML model, a prestored current medical literature to obtain a disease-specific medical literature analysis. The method may further include automatically generating at least one recommendation for the patient based on the adaptive trial platform, the disease-specific registry, the disease-specific medical literature analysis, and the real-time patient data in the EMR for providing real-time clinical decision support to the patient.
[006] In another embodiment, a system for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models is disclosed. The system may include a processor and a memory communicatively coupled to the processor. The memory may store processor-executable instructions, which, on execution, may cause the processor to receive real-time patient data corresponding to a patient. It should be noted that the real-time patient data may include geographical location of the patient, patient details, and information of a provider associated with the patient. The processor-executable instructions, on execution, may further cause the processor to associate an adaptive trial platform and a disease-specific registry of the patient with the EMR to monitor the real-time patient data. The processor-executable instructions, on execution, may further cause the processor to, upon associating, analyze a prestored current medical literature to obtain a disease-specific medical literature analysis. The processor-executable instructions, on execution, may further cause the processor to automatically generate at least one recommendation for the patient based on the adaptive trial platform, the disease-specific registry, the disease-specific medical literature analysis, and the real-time patient data in the EMR for providing real-time clinical decision support to the patient.
[007] In yet another embodiment, a non-transitory computer-readable medium storing computer-executable instruction for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models is disclosed. The stored instructions, when executed by a processor, may cause the processor to perform various operations including receiving real-time patient data corresponding to a patient. It should be noted that the real-time patient data may include geographical location of the patient, patient details, and information of a provider associated with the patient. The operations may further include associating an adaptive trial platform and a disease-specific registry of the patient with the EMR to monitor the real-time patient data. The operations may further include analyzing, through a first ML model, a prestored current medical literature to obtain a disease-specific medical literature analysis. The operations may further include automatically generating at least one recommendation for the patient based on the adaptive trial platform, the disease-specific registry, the disease-specific medical literature analysis, and the real-time patient data in the EMR for providing real-time clinical decision support to the patient.
[008] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS
[009] The present application can be best understood by reference to the following description taken in conjunction with the accompanying drawing figures, in which like parts may be referred to by like numerals
[010] FIG. 1 is a block diagram of a system for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models, in accordance with an embodiment.
[011] FIG. 2 is a flowchart of a method for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models, in accordance with an embodiment.
[012] FIG. 3 is a flowchart of a method for monitoring outcomes for one or more of the plurality of patients, in accordance with an embodiment.
[013] FIG. 4 is a block diagram of an exemplary computer system for implementing embodiments consistent with the present disclosure.

DETAILED DESCRIPTION OF THE DRAWINGS
[014] The following description is presented to enable a person of ordinary skill in the art to make and use the invention and is provided in the context of particular applications and their requirements. Various modifications to the embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the invention. Moreover, in the following description, numerous details are set forth for the purpose of explanation. However, one of ordinary skill in the art will realize that the invention might be practiced without the use of these specific details. In other instances, well-known structures and devices are shown in block diagram form in order not to obscure the description of the invention with unnecessary detail. Thus, the invention is not intended to be limited to the embodiments shown but is to be accorded the widest scope consistent with the principles and features disclosed herein.
[015] While the invention is described in terms of particular examples and illustrative figures, those of ordinary skill in the art will recognize that the invention is not limited to the examples or figures described. Those skilled in the art will recognize that the operations of the various embodiments may be implemented using hardware, software, firmware, or combinations thereof, as appropriate. For example, some processes can be carried out using processors or other digital circuitry under the control of software, firmware, or hard-wired logic. (the term “logic” herein refers to fixed hardware, programmable logic and/or an appropriate combination thereof, as would be recognized by one skilled in the art to carry out the recited functions.) Software and firmware can be stored on computer-readable storage media. Some other processes can be implemented using analog circuitry, as is well known to one of ordinary skill in the art. Additionally, memory or other storage, as well as communication components, may be employed in embodiments of the invention.
[016] Referring now to FIG. 1, a block diagram of a system 100 for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models is illustrated, in accordance with an embodiment. The system 100 may include a patient data analysis device 101. The patient data analysis device 101 may further include an association module 108, an analyzer 110, a recommendation generation module 114, a profile generation module 120, a monitoring module 122, and an alert generation module 124. Further, the patient data analysis device 101 may also include a database 112 to store various information and intermediate results generated by the modules 108-124. By way of an example, the modules 108-124 may perform various operations in order to analyze the patient data.
[017] The association module 108 may be configured to receive real-time patient data 102 corresponding to a patient. The real time patient data 102 may include personal information of a patient such as, but not limited to, a geographical location of the patient, patient details, and information of a provider associated with the patient. In an embodiment, the patient data may be collected from the patient through an input/output device such as a user device comprising, but not limited to, a laptop computer, a desktop computer, a notebook, a workstation, a portable computer, a personal digital assistant, a handheld device and a mobile device. In an embodiment, the data may be received from sources such as, but not limited to, patient’s insurance records, social media records, etc. based on patient’s consent and authentication.
[018] In an embodiment, the real-time patient data 102 may also include bio-medical data through various sensors attached to parts of the patient's body. The sensors may include bioelectrical signal transducers with either direct contact electrodes or indirect contact, e.g., capacitive or magnetic field electrodes. In an embodiment, the sensors may be physiological transducers, including, but not limited to, photoelectric, optical, acoustic, thermal, mechanical, chemical or radiation sensitive transducers. In an embodiment, the sensors may collect body function data, e.g., blood pressure, blood chemistry, heartbeat (arrhythmia), either transcutaneous using conventional non-invasive methods or intravenously.
[019] In an embodiment, the real-time patient data 102 may be utilized to generate a patient profile by a profile generation module 120. By way of an example, the patient details may include, but are not limited to, name, age, weight, height, address, unique identity, and other health indicators and physiological and medical data associated with the patient.
[020] In some embodiments, context may be added to the real-time patient data in the EMR. The context may be added based on context parameters generated based on historical data associated to patients’ medical history of a plurality of patients. In an embodiment, the context parameters may provide general information based on which the patient data may be aggregated. In an exemplary embodiment, each patient's hospital admission may be associated with a health condition or disease based information parameter requiring a hospitalization.
[021] The association module 108 may further be configured to process the real-time patient data based on information from an adaptive trial platform 104 and a disease-specific registry 106 of the EMR. The adaptive trial platform 104 may be configured to adaptively cumulate clinical trial data from various clinics, home care or hospitals. The clinical trial data may include trial data for any phase, in any therapeutic area, from around the world. The trial engine may be configured to generate and test study hypotheses in real time using data accumulated over time. The disease-specific registry 106 may include data corresponding to various prestored disease-specific medical data of various patients comprising historical medical data. Further, the disease-specific registry 106 may interface with literature from various sources, such as, but not limited to, medical journals, research databases, pharma laboratories, medical books, etc. for various diseases.
[022] The association module 108 may be configured to generate training data by associating the real-time patient data with one or more diseases based on the information from the adaptive trial platform 104 and the disease-specific registry 106. In an embodiment, the association module 108 may use various correlation models or machine learning algorithms for semantically correlating real-time patient data 102 with data from the adaptive trial platform 104 and the disease-specific registry 106.
[023] As illustrated in FIG. 1, the associated module 108 is communicatively coupled to the analyzer 110. The analyzer 110 may be configured for analyzing the output from the association module 108. The analyzer 110 may analyze the real-time patient data based on the training data provided by the association module using a first machine learning (ML) model 110A. In some embodiments, the first ML model 110A may be used for automatically reviewing the analyzed real-time patient data 102 and the diseases specific data and the trial information corresponding to the correlated diseases based on the training data. The first ML model may implement one or more machine learning techniques implemented by various algorithms such as, but not limited to, Probabilistic Graphical Models, Bayes Net, SVM, FT, Naïve Bayes, CNNs, ConvNets, etc.
[024] Further, in some embodiments, based on the analysis of the real-time patient data and the training data one or more treatment strategies may be determined. In an embodiment, the treatment strategies may include, suggestion of drugs, doses and schedule and method of delivery, the monitoring parameters such as critical and non-critical parameters and their threshold levels, etc. to be used for monitoring the patient. Further, the analyzer 110 may be operatively coupled to the database 112 which may save all the training data, analyzed data, etc.
[025] The analyzer 110 is coupled to the recommendation generation module 114. The recommendation generation module 114 may be configured to automatically generate at least one recommendation 116 for the patient. In an embodiment, the recommendation generated may be stored in the database 112. The recommendation 116 may be generated to depict various probability based scores related to various recommended parameters related to the recommended one or more treatment strategies. In an embodiment, the patient or a caretaker of the patient may agree to select one or more treatment plans based on the probability score associated with the risk parameters and other parameters of the recommended treatment plans. In an embodiment, the recommendation 116 may be regarding a real-time clinical decision support to the patient.
[026] Further, the recommendation generation module 114 may be operatively coupled to the profile generation module 120.
The profile generation module 120 may be configured to generate a profile for each of the plurality of patients. The profile may be generated based on the real-time patient data 102, the one or more recommendations 116 and patient outcomes 118 for each of the patients upon following the at least one recommendation. The profile generated for the patient may include details such as, but not limited to, patient information, recommendations given, outcomes, etc. and can be stored in the database 112 or transmitted to the monitoring module 122. For generating profiles, the profile generation module 120 may include a second ML model. The second ML model may be based on unsupervised machine learning algorithms.
[027] The monitoring module 122 may continuously monitor outcome(s) of each of the recommendations given to each of the plurality of patients. Further, the outcome(s) of the recommendations given to each of the plurality of patients with similar profiles are also monitored which were subjected to different treatments as per the recommendations.
[028] The patient outcomes 118 may be monitored based on real-time monitoring of the patient data 102 from the one or more sensors attached to the patient or one or more pharmacological results obtained from one or more pharmacological laboratories.
[029] In some embodiments, based on the continuous monitoring of the outcome(s) 118 from the patients, an unsupervised surveillance may be facilitated for research topics related to each medical condition. Moreover, an appropriate study design may be determined for a particular research question based on data obtained from the unsupervised surveillance.
[030] In an embodiment, while monitoring the patient outcomes 118 the monitoring module 122, in case the patient’s vital or monitoring parameters exceed the threshold levels may provide an alert to an alert generation module 124. The alert generation module 124 may in response generate an alert in the form of an alarm which may be sounded on patient’s data analysis device 101 or on a user device of the patient’s doctor or the caretaker. An alarm may be generated in the form of sounds and lights to capture the immediate attention of the doctor or caretaker. Further, the alert generation module 124 may be configured to generate an alert upon identifying a discrepancy in the outcome(s) 118 for the one or more of the plurality of patients.
[031] Therefore, the alarm generated may trigger the patient or caretaker or doctor to take remedial steps which may be generated in the form of new or alternative recommendations by the recommendation generation module 114 based on the machine learning module. The outcomes 118 of the new recommendation may be monitored in order to stabilize the patient’s vitals and to prevent the alarm from being triggered. Accordingly, the patient's profile would be updated based on new recommendations and outcomes by the profile generation module 120.
[032] The patient data analysis device 101 may allow for a continuous monitoring of the patients until the patients recover and provide them state of the art treatment plans and recommendations based on real-time monitoring and clinical data available throughout the treatment.
[033] It should be noted that the system 100 and associated patient data analysis device 101 may be implemented in programmable hardware devices such as programmable gate arrays, programmable array logic, programmable logic devices, or the like. Alternatively, the system 100 and the patient data analysis device 101 may be implemented in software for execution by various types of processors. An identified engine/module of executable code may, for instance, include one or more physical or logical blocks of computer instructions which may, for instance, be organized as an object, module, procedure, function, or other construct. Nevertheless, the executables of an identified engine/module need not be physically located together but may include disparate instructions stored in different locations which, when joined logically together, comprise the identified engine/module and achieve the stated purpose of the identified engine/module. Indeed, an engine or a module of executable code may be a single instruction, or many instructions, and may even be distributed over several different code segments, among different applications, and across several memory devices.
[034] As will be appreciated by one skilled in the art, a variety of processes may be employed to analyze patient data from the electronic medical record (EMR). For example, the exemplary system 100 and the patient data analysis device 101 may analyze the patient data from electronic medical record (EMR) through machine learning (ML) models, by the process discussed herein. In particular, as will be appreciated by those of ordinary skill in the art, control logic and/or automated routines for performing the techniques and steps described herein may be implemented by the system 100 and the patient data analysis device 101 either by hardware, software, or combinations of hardware and software. For example, suitable code may be accessed and executed by a processor in the patient data analysis device 101 to perform some or all of the techniques described herein. Similarly, application specific integrated circuits (ASICs) configured to perform some or all the processes described herein may be included in the processor in the patient data analysis device 101.
[035] Referring now to FIG. 2, a method for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models is depicted via a flowchart 200, in accordance with an embodiment. Each step of the method may be executed by the patient data analysis device 101. FIG. 2 is explained in conjunction with FIG. 1. It should be noted that each step of the method may be performed for each of a plurality of patients.
[036] At step 202, real-time patient data (such as real-time patient data 102) corresponding to a patient may be received. The real-time patient data may be received by an association module (same as the association module 108). The real-time patient data may include geographical location of the patient, patient details, and information of a provider associated with the patient. In some embodiments, context may be added to the real-time patient data in the EMR. The context may be added based on historical data associated with each of the plurality of patients. It should be noted that the patient details may include, but are not limited to, name, age, weight, height, address, unique identity, and other health indicators associated with the patient.
[037] Further, at step 204, an adaptive trial platform (same as the adaptive trial platform 104) and a disease-specific registry (same as the disease-specific registry 106) of the patient may be associated with the EMR. The association may be performed using the association module. Thus, the real-time patient data may be monitored.
[038] At step 206, a prestored current medical literature may be analyzed using an analyzer (same as the analyzer 110). The analysis may be performed through a first Machine Learning (ML) model (similar to the first ML model 110a of the analyzer 110). As a result, a disease-specific medical literature analysis may be obtained. In some embodiments, a current medical literature may be reviewed automatically through the first ML model. Hence, a type of disease and the recommended plan of action in each of the plurality of patients may be identified and suggested by the system based on the review. In some embodiments, a Bayesian clinical trial technique in association with the adaptive trial platform may be employed to analyze the real-time patient data.
[039] After that, at step 208, at least one recommendation (for example, the recommendation(s) 116) may be generated automatically for the patient. A recommendation module (same as the recommendation module 114) may be responsible for generating at least one recommendation. The recommendation may be generated based on the adaptive trial platform, the disease-specific registry, the disease-specific medical literature analysis, and the real-time patient data in the EMR. The one recommendation may help in providing a real-time clinical decision support to the patient. In an embodiment, the recommendation may provide a decision support in line or in consideration with any comorbidities associated to the patient. The recommendation may be draft notes on the EMR platform or draft orders on the same platform. The system will ask for the users’ approval by selection to execute the notes or orders.
[040] Referring now to FIG. 3, a method for monitoring outcomes for one or more of the plurality of patients is depicted via a flowchart 300, in accordance with an embodiment. FIG. 3 is explained in conjunction with Figs 1-2. At step 302, a profile for each of the plurality of patients may be generated using a profile generation module (similar to the profile generation module 120). In particular, a second ML model of the profile generation module may be used to generate the profile. The profile may be generated based on an outcome (same as the patient outcome(s) 118) for the patient upon following at least one recommendation. It should be noted that the second ML model may be based on unsupervised algorithms.
[041] Thereafter, at step 304, outcomes for one or more of the plurality of patients with similar profiles subjected to different treatments may be monitored continuously. To execute this step, a monitoring module (same as the monitoring module 122) may be used.
[042] At step 306, a discrepancy in the outcome may be monitored and in case no discrepancy is found in the outcomes then the process moves to step 304. In case a discrepancy is detected at step 306, the process may move to step 308 to generate an alert.
[043] In some embodiments, the process involving the trigger of an alarm at step 308 may move to step 310, where an unsupervised surveillance for suitable research topics may be facilitated. Further, at step 312, an appropriate study design may be determined for a particular research question based on data obtained from the unsupervised surveillance. It should be noted that the outcomes may be continuously monitored at step 304 after step 312.
[044] In some embodiments, based on the alert a new recommendation may be generated and based on which new outcomes may be generated at step 304 and the process may continue to continuously monitor the patient outcomes until successful treatment of the patient.
[045] In an embodiment, the ML model is updated based on the discrepancy detection, patient profile update based on continuous monitoring of the patient’s parameters and the recommendations generated. Accordingly, the ML model is adaptively trained over time with decisions made and consequences of the decisions seen. Thus, recommendations generated are as per the updated registries corresponding to the trail and diseases, outcome data and monitoring over time.
[046] The disclosed methods and systems may be implemented on a conventional or a general-purpose computer system, such as a personal computer (PC) or server computer. Referring now to FIG. 4, an exemplary computing system 400 that may be employed to implement processing functionality for various embodiments (e.g., as a SIMD device, client device, server device, one or more processors, or the like) is illustrated. Those skilled in the relevant art will also recognize how to implement the invention using other computer systems or architectures. The computing system 400 may represent, for example, a user device such as a desktop, a laptop, a mobile phone, personal entertainment device, DVR, and so on, or any other type of special or general-purpose computing device as may be desirable or appropriate for a given application or environment. The computing system 400 may include one or more processors, such as a processor 402 that may be implemented using a general or special purpose processing engine such as, for example, a microprocessor, microcontroller or other control logic. In this example, the processor 402 is connected to a bus 404 or other communication medium. In some embodiments, the processor 402 may be an Artificial Intelligence (AI) processor, which may be implemented as a Tensor Processing Unit (TPU), or a graphical processor unit, or a custom programmable solution Field-Programmable Gate Array (FPGA).
[047] The computing system 400 may also include a memory 406 (main memory), for example, Random Access Memory (RAM) or other dynamic memory, for storing information and instructions to be executed by the processor 402. The memory 406 also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed by the processor 402. The computing system 400 may likewise include a read only memory (“ROM”) or other static storage device coupled to bus 404 for storing static information and instructions for the processor 402.
[048] The computing system 400 may also include a storage device 408, which may include, for example, a media drive 410 and a removable storage interface. The media drive 410 may include a drive or other mechanism to support fixed or removable storage media, such as a hard disk drive, a floppy disk drive, a magnetic tape drive, an SD card port, a USB port, a micro USB, an optical disk drive, a CD or DVD drive (R or RW), or other removable or fixed media drive. A storage media 406 may include, for example, a hard disk, magnetic tape, flash drive, or other fixed or removable medium that is read by and written to by the media drive 420. As these examples illustrate, the storage media 412 may include a computer-readable storage medium having stored there in particular computer software or data.
[049] In alternative embodiments, the storage devices 408 may include other similar instrumentalities for allowing computer programs or other instructions or data to be loaded into the computing system 400. Such instrumentalities may include, for example, a removable storage unit 414 and a storage unit interface 416, such as a program cartridge and cartridge interface, a removable memory (for example, a flash memory or other removable memory module) and memory slot, and other removable storage units and interfaces that allow software and data to be transferred from the removable storage unit 414 to the computing system 400.
[050] The computing system 400 may also include a communications interface 418. The communications interface 418 may be used to allow software and data to be transferred between the computing system 400 and external devices. Examples of the communications interface 418 may include a network interface (such as an Ethernet or other NIC card), a communications port (such as for example, a USB port, a micro USB port), Near field Communication (NFC), etc. Software and data transferred via the communications interface 418 are in the form of signals which may be electronic, electromagnetic, optical, or other signals capable of being received by the communications interface 418. These signals are provided to the communications interface 418 via a channel 420. The channel 420 may carry signals and may be implemented using a wireless medium, wire or cable, fiber optics, or other communications medium. Some examples of the channel 420 may include a phone line, a cellular phone link, an RF link, a Bluetooth link, a network interface, a local or wide area network, and other communications channels.
[051] The computing system 400 may further include Input/Output (I/O) devices 422. Examples may include, but are not limited to a display, keypad, microphone, audio speakers, vibrating motor, LED lights, etc. The I/O devices 422 may receive input from a user and also display an output of the computation performed by the processor 402. In this document, the terms “computer program product” and “computer-readable medium” may be used generally to refer to media such as, for example, the memory 406, the storage devices 408, the removable storage unit 414, or signal(s) on the channel 420. These and other forms of computer-readable media may be involved in providing one or more sequences of one or more instructions to the processor 402 for execution. Such instructions, generally referred to as “computer program code” (which may be grouped in the form of computer programs or other groupings), when executed, enable the computing system 400 to perform features or functions of embodiments of the present invention.
[052] In an embodiment where the elements are implemented using software, the software may be stored in a computer-readable medium and loaded into the computing system 400 using, for example, the removable storage unit 414, the media drive 410 or the communications interface 418. The control logic (in this example, software instructions or computer program code), when executed by the processor 402, causes the processor 402 to perform the functions of the invention as described herein.
[053] Various embodiments provide methods and systems for analyzing patient data from electronic medical record (EMR) through machine learning (ML) models. The disclosed method and system may provide individualized treatment recommendations to clinicians leveraging existing medical data and new data that may be populated in the EMR. The recommendations may change as new evidence emerges without a usual delay associated with clinical trials.
[054] The disclosed system may offer incorporation of an adaptive trial engine and disease-based registries interfacing to an EMR. The system may create and analyze existing medical literature on specific diseases using machine learning algorithms. Further, the recommendations may be derived automatically from medical literature review. The recommendations may be based on the adaptive trial platform, the disease-specific registry, the disease-specific medical literature analysis, and the real-time patient data in the EMR which may be contextualized to geography and practice style (prior experience) of the provider.
[055] The outcomes may be continuously monitored for the patients with similar profiles (for example, end stage renal failure patients treated at tertiary care setting in Bihar) subjected to dissimilar treatments (for example, those who get high dose lasix prescriptions v/s those who do not). Further, an alert on any difference in outcomes may be generated. This may be performed using unsupervised ML algorithms such as a cluster analysis. This may help the appropriate audience to do surveillance for interesting research questions in an unsupervised fashion. Therefore, the patient data analysis device 102 may be used to identify the minimum effective dose (MED) and/or the maximum tolerable dose (MTD) using one or more of continuous monitoring using a reassessment method (CRM) or a Bayesian Logistic Regression Model (BLRM).
[056] In another embodiment, the patient data analysis device 102 may be used for an unsupervised analysis of outcomes based on recommendation generated. An appropriate study may be designed for a subset of patients and for identifying sub-populations with significantly higher response to treatment based on scientific rationales. Patients may undergo various testing such as molecular and/or pharmacological to determine eligibility for recommending a particular treatment or to enable interim decisions to continue with selected sub-populations or the full population based on the study.
[057] It will be appreciated that, for clarity purposes, the above description has described embodiments of the invention with reference to different functional units and processors. However, it will be apparent that any suitable distribution of functionality between different functional units, processors or domains may be used without detracting from the invention. For example, functionality illustrated to be performed by separate processors or controllers may be performed by the same processor or controller. Hence, references to specific functional units are only to be seen as references to suitable means for providing the described functionality, rather than indicative of a strict logical or physical structure or organization.
[058] Although the present invention has been described in connection with some embodiments, it is not intended to be limited to the specific form set forth herein. Rather, the scope of the present invention is limited only by the claims. Additionally, although a feature may appear to be described in connection with particular embodiments, one skilled in the art would recognize that various features of the described embodiments may be combined in accordance with the invention.
[059] Furthermore, although individually listed, a plurality of means, elements or process steps may be implemented by, for example, a single unit or processor. Additionally, although individual features may be included in different claims, these may possibly be advantageously combined, and the inclusion in different claims does not imply that a combination of features is not feasible and/or advantageous. Also, the inclusion of a feature in one category of claims does not imply a limitation to this category, but rather the feature may be equally applicable to other claim categories, as appropriate. 
, Claims:CLAIMS
I/We Claim:
1. A method for analysing patient data from electronic medical record (EMR) through machine learning (ML) models, the method comprising:
for each patient of a plurality of patients,
receiving, by a patient data analysis device (101), real-time patient data (102) corresponding to a patient, wherein the real-time patient data (102) comprises geographical location of the patient, patient details, and information of a provider associated with the patient;
associating, by the patient data analysis device (101), an adaptive trial platform (104) and a disease-specific registry (106) of the patient with the EMR to monitor the real-time patient data (102);
upon associating, analysing, by the patient data analysis device (101) and through a first ML model (110A), a prestored current medical literature to obtain a disease-specific medical literature analysis; and
automatically generating, by the patient data analysis device (101), at least one recommendation (116) for the patient based on the adaptive trial platform (104), the disease-specific registry (106), the disease-specific medical literature analysis, and the real-time patient data (102) in the EMR for providing a real-time clinical decision support to the patient.

2. The method of claim 1, comprises generating, through a second ML model, a profile for each of the plurality of patients based on an outcome for the patient upon following the at least one recommendation, wherein the second ML model is based on unsupervised algorithms.

3. The method of claim 2, comprises continuously monitoring outcomes for one or more of the plurality of patients with similar profiles subjected to different treatments.

4. The method of claim 3, comprises:
facilitating unsupervised surveillance for suitable research topics based on the continuously monitored outcomes; and
determining an appropriate study design for a particular research question based on data obtained from the unsupervised surveillance.

5. The method of claim 3, comprises generating an alert upon identifying a discrepancy in the outcomes for the one or more of the plurality of patients.

6. The method of claim 1, comprises automatically reviewing the current medical literature through the first ML model (110A) to identify a type of disease in each of the plurality of patients, wherein the current medical literature comprises a published medical literature.

7. The method of claim 1, comprises adding context to the real-time patient data (102) in the EMR based on historical data associated with each of the plurality of patients.

8. The method of claim 1, comprises employing a Bayesian clinical trial technique in association with the adaptive trial platform (104) to analyse the real-time patient data (102).

9. A system for analysing patient data from electronic medical record (EMR) through machine learning (ML) models, the device comprising:
a processor (402); and
a memory (406) communicatively coupled to the processor (402), wherein the memory (406) stores processor-executable instructions, which, on execution, causes the processor (402) to:
receive real-time patient data (102) corresponding to the patient, wherein the real-time patient data (102) comprises geographical location of the patient, patient details, and information of a provider associated with the patient;
associate an adaptive trial platform (104) and a disease-specific registry (106) of the patient with the EMR to monitor the real-time patient data (102);
upon associating, analyze prestored current medical literature to obtain a disease-specific medical literature analysis; and
automatically generate at least one recommendation for the patient based on the adaptive trial platform (104), the disease-specific registry (106), the disease-specific medical literature analysis, and the real-time patient data (102) in the EMR for providing real-time clinical decision support to the patient.

10. The system of claim 9, wherein the processor-executable instructions cause the processor to generate, through a second ML model, a profile for each of the plurality of patients based on an outcome for the patient upon following the at least one recommendation, wherein the second ML model is based on unsupervised algorithms.