Method and system for managing health data across multiple subsystems

Abstract

The invention relates to a computer-aided method and system for managing health data across multiple subsystems. More specifically, this disclosure relates to a system that detects, processes, and analyzes data changes in real time, thereby supporting clinical and operational decision-making.

Description

[0001] DESCRIPTION

[0002] METHOD AND SYSTEM FOR MANAGING HEALTH DATA ACROSS MULTIPLE SUBSYSTEMS

[0003] Technical Field of the Invention

[0004] The invention relates to a computer-aided method and system for managing health data across multiple subsystems. More specifically, this disclosure relates to a system that detects, processes, and analyzes data changes in real time, thereby supporting clinical and operational decision-making.

[0005] State of the Art of the Invention

[0006] In the healthcare sector, software developed to support hospital management processes is generally designed to increase efficiency, improve patient care, and optimize operational processes. Hospital Information Management System (HIMS), one of the most widely used among these softwares, is a fundamental tool for digitally storing and managing patient and operational data. However, HIMS falls short in terms of further analyzing and interpreting these data and integrating them into strategic decision-making processes. For example, current systems are not able to consolidate data from different sources, nor are they able to derive meaning from this data and optimize processes.

[0007] HIMS systems used in hospitals generally only fulfill the task of data collection and reporting, but do not allow these data to be used effectively in decision-making processes for the overall management of hospital operations. Subsystems where existing systems are not integrated or are integrated in a limited way are far from providing the holistic structure needed to support the wide-ranging processes of hospitals. For example, systems such as PACS and LIS only cover imaging and laboratory processes; they do not have the infrastructure to ensure that this data is meaningfully associated with other processes.

[0008] In addition to these shortcomings, while clinical decision support systems offer artificial intelligence and machine learning-based solutions in the healthcare sector, they are mostly focused on a specific medical field or diagnosis. Such systems are limited in contributing toclinical decision-making processes by analyzing patient anamnesis data and laboratory results in an integrated manner. For example, current artificial intelligence applications focus more on the analysis of radiological images, but are not effective enough in interpreting general patient data and supporting individualized treatment processes.

[0009] The inability to process data from different systems in an integrated manner directly affects the efficiency of operational processes. For example, problems such as failure to optimize operating room usage times, inefficient use of devices, and failure to take advantage of appointment gaps lead to both increased costs and decreased patient satisfaction. Furthermore, deficiencies in clinical processes can lead to increased time spent per patient and delays in diagnostic processes.

[0010] Today, generic database and business intelligence tools such as Oracle, Microsoft SQL Server and PostgreSQL meet the data management needs of a wide range of businesses. However, these tools fall short of providing customized solutions for organizations with specific needs, such as hospitals, and can often be implemented at high costs. In addition, some internationally developed clinical decision support systems focus only on specific medical applications and are unable to provide comprehensive solutions to support overall hospital management processes.

[0011] Compared to existing technologies, the advantages of the proposed solution are quite evident. Unlike HIMS and other systems that are limited in operational process optimization, the proposed digital management platform and clinical decision support system are designed directly for hospital management and healthcare industry. A comparison of existing systems and the proposed solution is summarized below:

[0012] The developed models are distinguished from existing HIMS and international systems by their ability to provide forward-looking predictions. Offering a healthcare-specific customized structure for the optimization of operational processes, this solution directly integrates specific know-how and analytical approaches, in contrast to existing systems which are often generic and costly. The system, which exhibits a design suitable for hospital management in terms of data analytics and integration processes, provides a more effective structure compared to existing systems with complex and costly integrations. Furthermore, by enabling consolidated data management while avoiding data loss through ETL and DWH integration,this approach contributes directly to clinical processes by integrating patient history data and laboratory results within clinical decision support systems, this is a much more comprehensive benefit than is offered by existing systems which often focus solely on radiological imaging.

[0013] All these limitations of existing technologies clearly demonstrate the need for more integrated, efficient and artificial intelligence-aided solutions in hospital management. In the digital transformation process of the healthcare sector, optimizing operational processes, processing data in a meaningful way and expanding clinical decision support systems are of great importance as solutions to complement the shortcomings of existing software.

[0014] As a result, all the above-mentioned problems have made it imperative to make an innovation in the relevant field.

[0015] Objects and Summary of the Invention

[0016] The object of the invention is to eliminate data loss and confusion caused by integration processes by enabling the consolidated collection of distributed data through ETL and DWH infrastructure, and to enable effective analysis and interpretation of data from different systems in hospitals.

[0017] Another object of the invention is to create a customized software architecture directly for hospital management processes and to build a structure that can access, process and, interpret all data in subsystems, not just data reporting.

[0018] Another object of the invention is to create a digital transformation platform in the healthcare sector by providing an infrastructure with near real-time data processing, BI (business intelligence), and command center features.

[0019] Another object of the invention is to provide clinical decision support systems using artificial intelligence-based algorithms to assist physicians.

[0020] In order to achieve the above objects, the present invention comprises the following steps: real-time detection of changes occurring in data across multiple source systems that digitallycontain patient medical records and information, appointment information, medication tracking information, and laboratory results, using a change data capture module; transmission of detected data changes to a message queuing system using a message broker, extraction of the transmitted data, conversion of the extracted data into a standard format suitable for downstream processing, and loading of the converted data into a data warehouse system; and analysis of the received data using predefined data processing algorithms to perform resource utilization optimization, patient flow prediction, and diagnostic support decisions.

[0021] Detailed Description of the Invention

[0022] The subject of the invention relates to a computer-aided method and system for managing health data across multiple subsystems. More specifically, this disclosure relates to a system that detects, processes, and analyzes data changes in real time, thereby supporting clinical and operational decision-making.

[0023] This disclosure provides a computer-aided method and system for managing health data across multiple subsystems. The system includes a change data capture (CDC) module that detects data changes in real time from various health information systems such as hospital information management systems (HIMS), laboratory information systems (LIS) and picture archiving and communication systems (PACS). Detected data changes are transmitted via a message broker to a message queuing system and routed to the appropriate processing components based on defined routing rules.

[0024] An ETL (Extract, Transform, Load) module extracts and transforms data and loads it into a data warehouse (DWH) system. The data warehouse system collects and stores data from multiple source systems and stores it in an optimized structure for analysis. The collected data is analyzed using defined data processing algorithms and insights are generated for clinical and operational decision-making processes from these analyses. These insights include optimizing resource utilization, predicting patient flow, and supporting diagnostic decisions. Optionally, it also provides recommendations for budget planning, cost analysis, and smart pricing.The change data capture module is responsible for detecting data changes in real time from various source systems such as hospital information management systems (HIMS), laboratory information systems (LIS), and picture archiving and communication systems (PACS).

[0025] The change data capture module works with other components, such as the message broker, to communicate detected data changes to the processing components. The message broker routes these changes according to defined routing rules and ensures that the data is delivered to the appropriate processing components. Preferably RabbitMQ is used as the message broker. RabbitMQ is an open source software used as a message broker or messaging tool. RabbitMQ supports the AMQP (Advanced Message Queuing Protocol) standard and can also use other protocols such as HTTP and MQTT. Furthermore, the message broker interacts with a message queuing system to manage the flow of data changes. Routing rules determine the path that data changes take and ensure that these changes reach the right processing components efficiently.

[0026] The message broker is configured to manage data changes from various source systems such as hospital information management systems (HIMS), laboratory information systems (LIS), and picture archiving and communication systems (PACS).

[0027] The ability of the message broker to effectively communicate and route data changes plays an important role in the overall functionality of the health data management system. This component supports the integrity and efficiency of the system by facilitating the seamless flow of data between various components and subsystems.

[0028] The ETL Module performs the functions of extracting, transforming, and loading data into a data warehouse system. Here, it receives data from the message broker and ensures that the data is captured for correct processing. It then standardizes and transforms this extracted data into a format suitable for downstream processing.

[0029] After the conversion process, the loader saves the standardized data in a data warehouse system. A data warehouse system is designed to collect and store data from multiple source systems. This consolidation provides a central repository of health data to support analysis and reporting. A data warehouse system stores data in an optimized structure, making it easier to retrieve and analyze data effectively. The ETL Module ensures that data is accuratelyprocessed and stored, enabling insights to be generated for clinical and operational decisionmaking.

[0030] A processing unit analyzes the collected data and generates insights. These components perform analysis using machine learning algorithms and data processing algorithms. The analysis process involves the application of defined data processing algorithms to generate insights to aid clinical and operational decision-making. This process encompasses functions such as optimizing resource utilization, predicting patient flow, and supporting diagnostic decisions.

[0031] Processing components play a critical role for healthcare operations by transforming raw data into actionable insights.

[0032] The insights generated by the processing components can be accessed through a user interface with real-time monitoring and reporting capabilities. This integration ensures a seamless flow of information and enables healthcare providers to make informed decisions based on the most up-to-date data.

[0033] The user interface component provides access to processed data and insights. This component includes a dashboard with real-time monitoring and reporting functions. The user interface provides access to insights derived from the collected data and allows users to interact with this data. Processed data can be accessed through this interface and these data enable visualization of clinical and operational metrics.

[0034] The user interface is designed to facilitate the presentation of data in a way that supports decision-making processes. Furthermore, the user interface offers customizable alerts and reports based on predefined clinical and operational thresholds. This customization feature allows users to tailor the interface to suit their needs and preferences. The user interface component serves as the primary point of interaction for users to utilize insights derived from the data management system. With these features, the user interface plays a critical role in the overall functionality of the system.

Claims

CLAIMS1. A computer-implemented method for managing health data across multiple subsystems, characterized in that it comprises:• real-time detection of changes occurring in data across multiple source systems that digitally contain patient medical records and information, appointment information, medication tracking information, and laboratory results, using a change data capture module;• transmission of detected data changes to a message queuing system using a message broker,• extraction of the transmitted data, conversion of the extracted data into a standard format suitable for downstream processing, and loading of the converted data into a data warehouse system,• analysis of the received data using predefined data processing algorithms to perform resource utilization optimization, patient flow prediction, and diagnostic support decisions.

2. The method according to claim 1, characterized in that a pre-trained machine learning algorithm is used to perform resource utilization optimization, patient flow prediction, and diagnostic support decisions.

3. The method according to claim 1, characterized in that said analysis generates recommendations for optimizing resource allocation, including staff planning, equipment utilization, and bed capacity.

4. The method according to claim 1, characterized in that said analysis generates recommendations for budget planning, cost analysis, and smart pricing.

5. The method according to claim 3 or 4, characterized in that said analysis is performed by regression analysis and an artificial neural network.

6. A method according to claim 1, characterized in that RabbitMQ is used as a message broker.

7. A data management system for managing healthcare operations across multiple subsystems, characterized in that it comprises:• a change data capture (CDC) module configured to detect data changes across multiple source systems in real time;• a message broker configured to communicate detected data changes to one or more processing components,• an ETL module that receives data from the message broker, standardizes the extracted data, and loads the transformed data into a data warehouse;• a data warehouse system, and• a processor unit for analyzing the received data using predefined data processing algorithms to perform resource utilization optimization, patient flow prediction, and diagnostic support decisions.