Presentation of patient summaries in ehr

Abstract

Described are techniques for summarizing patient history, diagnoses, input from other medical systems, etc., and generating output in various summary formats: visual UI, written or verbalized text of various lengths, and key elements to guide a prediction engine. It has three components: a data integration engine, a real-time intelligent data analytics engine, and a summarization and representation engine. Inputs can be from a health information exchange (HIE), which can include a number of different electronic health records (EHR) or other sources of health-related data; wearables; and personal health records (PHR). Outputs can include summaries at various hierarchical levels of time and complexity, and can include text, audio, visual, tabular, and other formats, and which can be used for predictions.

Description

PRESENTATION OF PATIENT SUMMARIES IN EHRBACKGROUND

[0001] A patient generates significant amounts of data over a lifetime. Healthcare providers, facilities, and systems require access to that data to obtain an overview before consulting with the patient or to identify problems, develop treatment plans, and provide care to the patient. Clinical decision making is complicated by the overwhelming amount of data and the limited time able to be spent with each patient. Being able to peruse the data from different sources in a usable format prior to meeting with the patient will improve efficiency and overall patient care.SUMMARY

[0002] This system summarizes patient history, diagnoses, input from other medical systems, and generates output in various summary formats: visual UI, written or verbalized text of various lengths, and key elements to guide a prediction engine. It has three components: a data integration engine, a real-time intelligent data analytics engine, and a summarization and representation engine.

[0003] Inputs can be from a health information exchange (HIE), which can include a number of different EHRs or other sources of health-related data; wearables; personal health records (PHR); and current and past data from a CharmHealth datastore. Outputs can include summaries at various hierarchical levels of time and complexity, and can include text, audio, visual, tabular, and other formats, and which can be used for predictions.

[0004] The data integration engine collates the data from several sources including past EHR data, wearables, patient input forms and questionnaires, and data from HIE / other EHRs from clinics the patient may have visited.

[0005] The intelligent data analytics engine digests patient information, complaints, changes, diagnosis and procedure codes, referrals, insurance, billing and other key points. It implements algorithms to extract keywords and key features to generate the outputs. It can also integrate wearable health device data for proactive care management.

[0006] The summarization and representation engine provides an audio and / or visual overview of the patient’s data for quick assessment. The visual overview is typically a user- friendly display of salient points in the patient’s health history. It can be a hierarchicalrepresentation that can navigate to the source EHR data. The audio overview uses advanced Al techniques to highlight important details in a natural language narrative, addressing time constraints. This can help the physician or care provider listen to or read the summary on a mobile device.

[0007] The engine can also consider referrals and allows organizations to calculate metrics for value-based care organizations. It can generate a comprehensive patient history of different levels of detail promoting compliance and transparency.

[0008] This system can provide input to a predictive engine that can perform three types of prediction: accurate insurance billing codes (ICD and procedure), future treatment planning, and future patient outcomes. It saves time and ensures proper coding for insurance.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG. 1 depicts a diagram of an example of data source summarization.DETAILED DESCRIPTION

[0010] This is a system designed to provide medical professionals with a comprehensive summary of a patient's history in various audio / visual formats.

[0011] FIG. 1 depicts a diagram 100 of an example of data source summarization. The diagram 100 includes a data integration engine 102, a real-time intelligent data analytics engine 104 coupled to the data integration engine 102, and a summarization and representation engine 106 coupled to the real-time intelligent data analytics engine 104. The diagram 100 depicts examples of data sources to the data integration engine 102, including HIE / other EHRs, CharmHealth, labs, wearables, and clinical trials. The diagram 100 depicts examples of output from the summarization and representation engine 106, including collated and ranked data and data predictions.

[0012] The diagram 100 includes optional components that make use of output from the summarization and representation engine 106, including a UI 108, text / audio 110, and a code prediction engine 112. The code prediction engine can provide questions to ask patients to ensure appropriate data is available for codes and international classification of diseases (ICD) and prediction codes. In a specific implementation, the code prediction engine 112 is coupled to an insurance billing engine (not shown).

[0013] The components illustrated in diagram 100 can be connected via a network. The network and other networks discussed in this paper are intended to include all communication paths that are statutory (e.g., in the United States, under 35 U.S.C. 101), and to specifically exclude all communication paths that are non-statutory in nature to the extent that the exclusion is necessary for a claim that includes the communication path to be valid. Known statutory communication paths include hardware (e.g., registers, random access memory (RAM), nonvolatile (NV) storage, to name a few), but may or may not be limited to hardware.

[0014] The network and other communication paths discussed in this paper are intended to represent a variety of potentially applicable technologies. For example, the network can be used to form a network or part of a network. Where two components are co-located on a device, the network can include a bus or other data conduit or plane. Where a first component is co-located on one device and a second component is located on a different device, the network can include a wireless or wired back-end network or LAN. The network can also encompass a relevant portion of a WAN or other network, if applicable.

[0015] The devices, systems, and communication paths described in this paper can be implemented as a computer system or parts of a computer system or a plurality of computer systems. In general, a computer system will include a processor, memory, non-volatile storage, and an interface. A typical computer system will usually include at least a processor, memory, and a device (e.g., a bus) coupling the memory to the processor. The processor can be, for example, a general-purpose central processing unit (CPU), such as a microprocessor, or a special-purpose processor, such as a microcontroller.

[0016] The memory can include, by way of example but not limitation, random access memory (RAM), such as dynamic RAM (DRAM) and static RAM (SRAM). The memory can be local, remote, or distributed. The bus can also couple the processor to non-volatile storage. The nonvolatile storage is often a magnetic floppy or hard disk, a magnetic-optical disk, an optical disk, a read-only memory (ROM), such as a CD-ROM, EPROM, or EEPROM, a magnetic or optical card, or another form of storage for large amounts of data. Some of this data is often written, by a direct memory access process, into memory during execution of software on the computer system. The non-volatile storage can be local, remote, or distributed. The non-volatile storage is optional because systems can be created with all applicable data available in memory.

[0017] Software is typically stored in the non-volatile storage. Indeed, for large programs, it may not even be possible to store the entire program in the memory. Nevertheless, it should beunderstood that for software to run, if necessary, it is moved to a computer-readable location appropriate for processing, and for illustrative purposes, that location is referred to as the memory in this paper. Even when software is moved to the memory for execution, the processor will typically make use of hardware registers to store values associated with the software, and local cache that, ideally, serves to speed up execution. As used herein, a software program is assumed to be stored at an applicable known or convenient location (from non-volatile storage to hardware registers) when the software program is referred to as "implemented in a computer- readable storage medium." A processor is considered to be "configured to execute a program" when at least one value associated with the program is stored in a register readable by the processor.

[0018] In one example of operation, a computer system can be controlled by operating system software, which is a software program that includes a file management system, such as a disk operating system. One example of operating system software with associated file management system software is the family of operating systems known as Windows® from Microsoft Corporation of Redmond, Washington, and their associated file management systems. Another example of operating system software with its associated file management system software is the Linux operating system and its associated file management system. The file management system is typically stored in the non-volatile storage and causes the processor to execute the various acts required by the operating system to input and output data and to store data in the memory, including storing files on the non-volatile storage.

[0019] The bus can also couple the processor to the interface. The interface can include one or more input and / or output (I / O) devices. Depending upon implementation-specific or other considerations, the EO devices can include, by way of example but not limitation, a keyboard, a mouse or other pointing device, disk drives, printers, a scanner, and other EO devices, including a display device. The display device can include, by way of example but not limitation, a cathode ray tube (CRT), liquid crystal display (LCD), or some other applicable known or convenient display device. The interface can include one or more of a modem or network interface. It will be appreciated that a modem or network interface can be considered to be part of the computer system. The interface can include an analog modem, ISDN modem, cable modem, token ring interface, satellite transmission interface (e.g., "direct PC"), or other interfaces for coupling a computer system to other computer systems. Interfaces enable computer systems and other devices to be coupled together in a network.

[0020] The computer systems can be compatible with or implemented as part of or through a cloud-based computing system. As used in this paper, a cloud-based computing system is a system that provides virtualized computing resources, software and / or information to end user devices. The computing resources, software and / or information can be virtualized by maintaining centralized services and resources that the edge devices can access over a communication interface, such as a network. "Cloud" may be a marketing term and for the purposes of this paper can include any of the networks described herein. The cloud-based computing system can involve a subscription for services or use a utility pricing model. Users can access the protocols of the cloud-based computing system through a web browser or other container application located on their end user device.

[0021] A database management system (DBMS) can be used to manage a datastore. In such a case, the DBMS may be thought of as part of the datastore, as part of a server, and / or as a separate system. A DBMS is typically implemented as an engine that controls organization, storage, management, and retrieval of data in a database. DBMSs frequently provide the ability to query, backup and replicate, enforce rules, provide security, do computation, perform change and access logging, and automate optimization. Examples of DBMSs include Alpha Five, DataBase, Oracle database, IBM DB2, Adaptive Server Enterprise, FileMaker, Firebird, Ingres, Informix, Mark Logic, Microsoft Access, InterSystems Cache, Microsoft SQL Server, Microsoft Visual FoxPro, MonetDB, MySQL, PostgreSQL, Progress, SQLite, Teradata, CSQL, OpenLink Virtuoso, Daffodil DB, and OpenOffice.org Base, to name several.

[0022] Database servers can store databases, as well as the DBMS and related engines. Any of the repositories described in this paper could presumably be implemented as database servers. It should be noted that there are two logical views of data in a database, the logical (external) view and the physical (internal) view. In this paper, the logical view is generally assumed to be data found in a report, while the physical view is the data stored in a physical storage medium and available to a specifically programmed processor. With most DBMS implementations, there is one physical view and an almost unlimited number of logical views for the same data.

[0023] A DBMS typically includes a modeling language, data structure, database query language, and transaction mechanism. The modeling language is used to define the schema of each database in the DBMS, according to the database model, which may include a hierarchical model, network model, relational model, object model, or some other applicable known or convenient organization. An optimal structure may vary depending upon application requirements (e.g., speed, reliability, maintainability, scalability, and cost). One of the morecommon models in use today is the ad hoc model embedded in SQL. Data structures can include fields, records, files, objects, and any other applicable known or convenient structures for storing data. A database query language can enable users to query databases and can include report writers and security mechanisms to prevent unauthorized access. A database transaction mechanism ideally ensures data integrity, even during concurrent user accesses, with fault tolerance. DBMSs can also include a metadata repository; metadata is data that describes other data.

[0024] As used in this paper, a data structure is associated with a particular way of storing and organizing data in a computer so that it can be used efficiently within a given context. Data structures are generally based on the ability of a computer to fetch and store data at any place in its memory, specified by an address, a bit string that can be itself stored in memory and manipulated by the program. Thus, some data structures are based on computing the addresses of data items with arithmetic operations; while other data structures are based on storing addresses of data items within the structure itself. Many data structures use both principles, sometimes combined in non-trivial ways. The implementation of a data structure usually entails writing a set of procedures that create and manipulate instances of that structure. The datastores, described in this paper, can be cloud-based datastores. A cloud-based datastore is a datastore that is compatible with cloud-based computing systems and engines.

[0025] A computer system can be implemented as an engine, as part of an engine or through multiple engines. As used in this paper, an engine includes one or more processors or a portion thereof. A portion of one or more processors can include some portion of hardware less than all the hardware comprising any given one or more processors, such as a subset of registers, the portion of the processor dedicated to one or more threads of a multi -threaded processor, a time slice during which the processor is wholly or partially dedicated to carrying out part of the engine's functionality, or the like. As such, a first engine and a second engine can have one or more dedicated processors or a first engine and a second engine can share one or more processors with one another or other engines. Depending upon implementation-specific or other considerations, an engine can be centralized or its functionality distributed. An engine can include hardware, firmware, or software embodied in a computer-readable medium for execution by the processor that is a component of the engine. The processor transforms data into new data using implemented data structures and methods, such as is described with reference to the figures in this paper.

[0026] The engines described in this paper, or the engines through which the systems and devices described in this paper can be implemented, can be cloud-based engines. As used in this paper, a cloud-based engine is an engine that can run applications and / or functionalities using a cloud-based computing system. All or portions of the applications and / or functionalities can be distributed across multiple computing devices and need not be restricted to only one computing device. In some embodiments, the cloud-based engines can execute functionalities and / or modules that end users access through a web browser or container application without having the functionalities and / or modules installed locally on the end-users' computing devices.

[0027] Referring once again to the example of FIG. 1, the data integration engine 102 digests two types of data: individual data and population level data. Individual data includes a patient's chart and complete history, with special emphasis given to recent major events, significant complaints, notable changes, referrals, and other key points that may impact the patient's current health condition. Individual data also includes data imported via a HIE from other medical centers through third party integrations or HL7. Individual data also includes data from wearable health devices and sensors. Population level data includes aggregated, de-identified data from multiple patients, including many or all the individual data types. Population level data also can include clinical trial data, public datasets, and epidemiological data. This integration enables medical professionals to monitor important health indicators and identify trends for proactive care management.

[0028] The real-time intelligent data analytics engine 104 implements data processing to extract the salient features and keywords and structured data from the input. In a specific implementation, the real-time intelligent data analytics engine 104 uses machine learning (ML) algorithms and heuristics to extract a weighted representation of patient data. Weighted representations can provide special emphasis to recent major events, significant complaints, notable changes, and other key points that may impact the patient's current health condition. This component also considers patient referrals, ensuring that all relevant information may be included in the summary. Additionally, accountable care organizations can leverage the system's clinical summaries to calculate important metrics, supporting effective care coordination and management. This component would provide unique summaries of the realtime intelligent data analytics engine 104 based on the role of the clinical provider in the patient’s network of care.

[0029] With an advanced clinical summary representation, the summarization and representation engine 108 offers an audio and / or visual overview of the patient'scomprehensive data, aiding in quick and efficient assessment of the patient's health status. In a specific implementation, the visual overview has a summary presentation including key elements in one screen. It is presented for a quick one-glance summary before meeting with the patient. The visual summary can include graphical tools such as WordClouds emphasizing key elements in larger font, a bubble chart representing diagnoses / symptoms or other salient features, etc.

[0030] Using advanced natural language processing techniques and generative Al, the audio overview aspect takes the data from the real-time intelligent data analytics engine 104, highlighting the most salient parts in a concise and informative natural language (NL) description. Furthermore, it addresses the time constraints faced by doctors by presenting a verbal description of the patient's history in a given length of time, such as five minutes. This allows medical professionals to listen to the patient's history while multitasking, such as during preparation or while driving.

[0031] In a specific implementation, an algorithm employed by the summarization and representation engine 106 utilizes a multi-faceted approach to generate comprehensive clinical summaries. The algorithm considers various factors and sources of information, employing advanced natural language processing (NLP) techniques and generative artificial intelligence (Al) models. It prioritizes key data from the real-time intelligent data analytics engine 104 and assigns weights to different components based on relevance, significance, and impact on patient health status.

[0032] Summarization and representation can include the following steps:

[0033] / . Data Collection and Pre-processing

[0034] 1.1. Gather patient data from various electronic health record (EHR) systems — including medical history, diagnostic reports, treatment plans, and relevant notes — and population-level data.

[0035] 1.2. Pre-process the data to clean and structure it for analysis.

[0036] 2. Salience Scoring

[0037] 2.1. Apply salience scoring to different components of the patient's medical and familial history based on factors such as recency, severity, and relevance to the current health condition.

[0038] 2.2. Assign weights to events like recent major health events, significant complaints, and notable changes.

[0039] 2.3. Assign comparative values and metrics to situate the patient's unique health trajectory in the population at large.

[0040] 3. Natural Language Generation

[0041] 3.1. Utilize generative Al models to convert salient data points into a coherent and concise natural language description.

[0042] 3.2. Ensure the generated text aligns with the user's context and preferences.

[0043] 4. Time or Complexity Constrained Verbal Summary

[0044] 4.1. Implement algorithms to condense the information into a verbal summary suitable for a specific time frame or complexity metric, such as a five-minute overview or a simplelanguage representation.

[0045] 4.2. Prioritize information based on its importance and impact on the patient's health.

[0046] 5. Role-Based Customization

[0047] 5.1. Customize the summary based on the role of the clinical provider in the patient's network of care.

[0048] 5.2. Highlight information that is most relevant to the specific responsibilities and interests of the healthcare professional who is receiving the summary.

[0049] For example, consider a scenario where a doctor is preparing to meet a patient for a routine checkup. The summarization and representation engine, using the described algorithm, provides a concise and informative audio overview. The algorithm assigns higher weights to recent major events, notable changes, and significant complaints. The generated verbal summary, within a specified time frame, emphasizes these salient points, ensuring the doctor is well-informed about crucial aspects of the patient's health before the appointment.

[0050] Components are weighted and sorted by salience score (consider recent major symptoms, emergency department admissions, urgent care visits, notable changes in symptoms, significant complaints); Patient Referrals; Family History; Other Key Points.

[0051] The algorithm ensures that the most impactful information is presented prominently, allowing for an efficient assessment of the patient's health status during the limited time available for the clinical encounter.

[0052] The data from this system can be used by the business analytics system of the medical group in many ways. The data can be aggregated across all the facilities, clinics, providers orpatient cohorts to determine various metrics and analytics such as most encountered diagnosis codes, and procedure codes; top prescriptions used for each condition and their outcomes / efficacy; business revenue analysis across groups, patient churn, recurring revenue models; and labs and tests, and the use of various biomarkers or biomarker candidates, etc.

[0053] The advanced clinical summary generated by the summarization and representation engine is valuable for insurance companies. It helps them analyze patient outcomes and healthcare professionals' performance. The system provides comprehensive summaries with key indicators such as patient complaints, treatment adherence, and the success of medical interventions. It also allows insurance companies to evaluate the effectiveness of healthcare teams and individual providers. These metrics help them make data-driven decisions, optimize healthcare outcomes, and allocate resources more effectively.

[0054] The detailed clinical summaries generated by the summarization and representation engine component can also be input into a predictive engine. A predictive engine can utilize the available information to perform 3 forms of prediction: predict accurate ICD codes for insurance billing; predict accurate procedure codes for insurance billing; and predict future patient outcomes (diagnosis and prognosis).

[0055] By automating this process, medical professionals can save time and ensure proper coding for insurance purposes.

[0056] Doctors do not have time to go through a full timeline; everything must be on the timeline, which makes it even more difficult to navigate for important parts. Techniques described in this paper allow for converting the timeline into a summary and providing the summary of a patient’s recent medical history in multiple formats so a physician can quickly, at one glance, digest key information, before meeting the patient or making a treatment decision.

[0057] A natural language (NL) description of a patient history in a given length, words, time, etc., e.g., NL description of patient history in 5 minutes, can be particularly useful for busy professionals. Use case: listen while prepping, while driving, etc. Techniques described in this paper enable digestion of chart and patient history and summarization of salient parts in NL. Emphasis is provided for more recent, major events, major complaints, changes, other key points. Rank is based on recency, seriousness, external (e.g., need to inform patient if drug is recalled).

[0058] Accountable care organizations need metrics. Clinical summaries can be used to calculate metrics (e.g., summary of summaries). One can also summarize input from wearables, summarize patient referrals, and create history for audit.

[0059] Summaries can be used by the business analytics software in the medical group to determine scores for efficacy, productivity, and value-based care metrics. Summary outputs can be used by insurance, auditing, and other third-party agencies to determine the performance of the medical group as a whole. The summary can be fed into additional processing engines, for example into predictive engines to predict future outcomes, generate relevant ICD codes, and generate procedure codes for billing and insurance purposes.

[0060] Generating ICD codes for billing insurance and other purposes. Insurance codes are so important, insurance coders are sometimes utilized. This can be replaced with a code prediction engine.

Claims

CLAIMSWhat is claimed is:

1. A system comprising: a data integration engine; an intelligent analytics engine coupled to the data integration engine; a summarization and representation engine coupled to the intelligent analytics engine; wherein, in operation, the data integration engine integrates personal health data associated with a patient and population health data from a health information exchange (HIE), the intelligent analytics engine computes a customizable hierarchical summary, and the summarization and representation engine presents the customizable hierarchical summary in a format that includes user interface (UI) components.

2. The system of claim 1, wherein the HIE includes an electronic health records (EHR) and a practice management system.

3. The system of claim 1, wherein the intelligent analytics engine is a real-time intelligent analytics engine.

4. The system of claim 1, wherein the intelligent analytics engine generates the customizable hierarchical summary in multiple formats.

5. The system of claim 1, wherein the UI components include a word cloud or bubble chart.

6. The system of claim 1, wherein the UI components include a tabular display.

7. The system of claim 1, wherein the UI components include audio or textual output for listening or reading.

Images