Abstract
Introduction Accurate diagnosis is essential for effective treatment in healthcare settings. Digital technologies have revolutionized medical diagnostics, particularly in pathology laboratory report analyses, by enhancing speed, quality, and precision. Diagnostic pathology has advanced significantly with digital imaging, AI algorithms, and computer-assisted techniques. Traditional machine learning methods require extensive training data, struggle with decision-making flexibility, and lack interpretability when handling new data. Methods This paper introduces a novel approach to address these challenges: a fuzzy logic-based system for interpreting pathology laboratory reports and diagnosing diseases, while considering the restrictions of engineered interpretations so that the degree of reliability of the output of the system is at par with human perception. Results The analysis highlights the impressive accuracy of the fuzzy logic-based diagnosis system, which is closely aligned with professional diagnoses. Notably, it correctly identified "normal" with an 83% probability, showcasing its potential for early disease detection. Performance evaluations indicated that the precision, recall, and F-measure significantly improved as the number of probable diagnoses considered increased. Furthermore, the system exhibited enhanced predictive accuracy for disease recurrence when incorporating multiple probable diagnoses, underscoring its robust clinical effectiveness. Conclusion A fuzzy logic-based system within an AI framework for interpreting pathology laboratory reports and diagnosing 17 diseases across various demographics is presented. The performance of the algorithm, assessed using specialized classification metrics, showed stable and satisfactory results. The analysis revealed a direct correlation between the number of probable diagnoses considered and improved diagnostic accuracy with enhancements in precision, recall, and F-measure. Broadening the diagnostic scope optimizes immediate patient care and long-term treatment planning, potentially enhancing patient outcomes and healthcare quality.