Interpretation of pathophysiology by laboratory data (3). A consultation program based on clinical laboratory data

Medical professionals often rely on clinical laboratory data to support diagnoses. While traditional methods focus on individual test results, integrating multiple data points into a coherent interpretation remains a challenge. Prior research has shown that isolated lab values may not fully reflect a patient's condition. That uncertainty drove the development of systems to synthesize lab data into broader diagnostic insights. No prior work had resolved how to combine lab results with expert knowledge in a structured way. The need for a tool that could process lab data and suggest possible diagnoses became clear. This gap motivated the creation of a consultation program using standardized metrics. The goal was to provide a framework for interpreting lab data in a way that aligns with clinical reasoning. This approach aimed to improve the consistency and usefulness of lab data in clinical settings.

The aim of this work was to develop a consultation program that interprets clinical laboratory data using a structured method. The program integrates multiple lab values into a diagnostic score for possible diseases. The motivation came from the need to support clinicians with a tool that can process complex data. The researchers proposed using standard deviation indices to normalize lab results. They also introduced weighting factors to reflect the importance of each test for specific diseases. The system ranks diseases based on calculated scores. This approach allows for a more systematic interpretation of lab data. The researchers intended to test this method in real clinical scenarios.

The consultation program converts lab data into standard deviation indices (SDI) to normalize values. Each SDI is then multiplied by a weighting factor (WF) assigned by medical experts. These WFs represent the relative importance of each test for a given disease. The program calculates a score for each disease by integrating the SDI-WF products. The top ten diseases with the highest scores are listed in descending order. Additional comments are included to provide context for the results. The system uses a modified Bayesian approach based on fuzzy set theory. This method replaces traditional probability with membership grades for disease likelihood.

The program was tested in nine patients, with evaluations performed 2-3 times during their clinical course. The results showed qualitative agreement with the actual clinical course and diagnosis in all cases. The system correctly ranked the most likely diseases according to the patients' conditions. The use of SDI and WF provided a consistent method for interpreting lab data. The fuzzy set approach allowed for nuanced disease probability estimates. The program's output included not only disease rankings but also supporting comments. These comments helped clinicians understand the reasoning behind the scores. The system demonstrated potential for improving diagnostic accuracy through structured data interpretation.

The consultation program based on SDI and WF provides a structured way to interpret lab data. The researchers propose that this method can support clinicians in making more informed decisions. The qualitative agreement with clinical outcomes suggests the system's usefulness. The fuzzy set approach allows for flexible disease probability assessments. The program's output includes both numerical scores and explanatory comments. The researchers suggest further testing in larger patient populations. They plan to conduct a quantitative evaluation using multiple cases in specific fields. This approach may lead to more robust validation of the system's diagnostic support capabilities.