Decision Support System Based on FHIR Profiles.

In several healthcare systems, patients often visit diagnostic centers without physician guidance. This is observed in about 28% of cases in Russia, where individuals seek laboratory tests independently. Such practices leave patients without professional assistance in interpreting diagnostic results. Manual interpretation is costly and time-consuming, which can delay patient care. Prior research has shown that automated systems can manage this task efficiently. However, no prior work had resolved how to apply these systems specifically for self-referred patients. This gap motivated the development of a system to support patient decision-making. No prior work had demonstrated how to translate raw test data into diagnostic recommendations for non-physician users. The absence of such a system created a need for a solution that could bridge this gap.

The study aimed to develop a decision support system for patients who visit diagnostic centers without a physician's referral. The system is designed to interpret laboratory test results and provide actionable recommendations. The motivation stems from the high percentage of patients in Russia who bypass traditional medical consultation. The goal is to reduce the burden on healthcare professionals and improve patient understanding of their results. The system must classify test results into potential diagnoses. It must also generate recommendations for each diagnosis. The challenge is to ensure the system is accurate and user-friendly for non-medical individuals. The study focuses on creating a scalable solution that can be adapted to different diagnostic contexts.

The researchers implemented a decision support system using clinical data and classification algorithms. The system processes a vector of laboratory test results and maps it to a set of possible diagnoses. The approach involves relating each test result to a predefined set of conditions. The system uses FHIR profiles to structure and exchange health data. The decision-making process is based on a classification model that identifies relevant diagnoses. Recommendations are generated for each diagnosis identified. The system is designed to be accessible to patients without medical training. The implementation includes validation steps to ensure the accuracy of generated reports.

The system successfully classified laboratory test results into potential diagnoses. It generated recommendations for each diagnosis based on the patient's test data. The classification model demonstrated high accuracy in matching test results to conditions. The system's output included both diagnostic suggestions and actionable advice. The use of FHIR profiles ensured compatibility with existing health data systems. The system's design allowed for easy integration into laboratory workflows. Patients received structured reports that explained their results in layman's terms. The implementation showed that automated systems can support patient decision-making effectively.

The study demonstrated that a decision support system can interpret laboratory test results for self-referred patients. The system uses FHIR profiles to structure data and generate reports. The authors propose that such systems can reduce the need for physician intervention in diagnostics. The results suggest that automated systems can provide accurate diagnostic classifications. The system's recommendations are based on established diagnostic criteria. The authors suggest that this approach can be adapted to other healthcare contexts. The study supports the idea that decision support systems can improve patient understanding of test results. The findings indicate that these systems can be integrated into existing diagnostic workflows.