Key Factor Detection for Anemia Prediction Using Fuzzy Inference Systems

Expert systems and fuzzy logic techniques are widely used for improving prediction accuracy by identifying the most influential factors. In the context of anemia prediction, it is essential to prioritize factors that contribute significantly to diagnosis. This study aims to assess and rank the importance of various factors influencing anemia prediction using advanced fuzzy logic approaches. Three different fuzzy logic techniques were employed to evaluate the significance of contributing factors for anemia prediction: the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), the Fuzzy Analytic Hierarchy Process (F -AHP), and an improved version of F -AHP that incorporates the geometric mean. These methods were used to systematically rank and assess the relevance of several parameters related to anemia prediction. The application of the F -AHP with geometric mean improvement provided a robust framework for decision-making, ensuring accurate and reliable ranking of factors. TOPSIS further enhanced the precision of parameter prioritization by comparing alternatives based on their closeness to the ideal solution. The analysis revealed that the Mean Corpuscular Volume (MCV) factor had the most significant impact on predicting anemia. The integrated fuzzy logic techniques, particularly the combination of F-AHP and its geometric mean improvement, proved effective in identifying key factors for anemia prediction. The M CV factor emerged as the most critical parameter, offering valuable insights for more accurate and reliable medical diagnoses.