Incorporating Confidence of Evidence in Diabetes Diagnosis Using Disc T-Spherical Fuzzy Sets with AHP–TOPSIS Framework

Wafa Alagal^1,*, Zanyar A. Ameen^2,*
1 Department of Mathematics and Statistics, College of Science, University of Jeddah, Jeddah, Saudi Arabia
2 Department of Mathematics, College of Science, University of Duhok, Duhok, Iraq
* Corresponding Author: Wafa Alagal. Email: email ; Zanyar A. Ameen. Email: email

Computer Modeling in Engineering & Sciences https://doi.org/10.32604/cmes.2026.083259

Received 31 March 2026; Accepted 14 May 2026; Published online 02 June 2026

Download PDF

Abstract

Diabetes remains a major global health challenge and requires diagnostic systems capable of handling uncertainty and sometimes conflicting clinical evidence. In this study, a Disc T-Spherical Fuzzy (DT-SF) TOPSIS framework is proposed for diabetes risk assessment, where the radius parameter is used to encode the confidence associated with each diagnostic attribute. The methodology also integrates the Analytic Hierarchy Process (AHP) to determine the relative importance of several key risk factors, including blood glucose, body mass index, family history, lifestyle factors, and clinical symptoms. One important feature of the proposed approach is the ternary classification scheme, which categorizes patients as Non-diabetic (N), Prediabetic (P), or Diabetic (D). In particular, this scheme allows the explicit identification of patients located in a grey zone (Class P), where early monitoring and preventive intervention may be beneficial. The proposed framework is evaluated using the Pima Indians Diabetes Dataset (PIDD). The obtained results show that the ternary DT-SF TOPSIS model achieves 89.14% accuracy, while the conventional binary thresholding method reaches 75.91%. Further analysis of the Closeness Coefficient (CC) distributions, together with threshold sensitivity examination, supports the robustness and interpretability of the proposed framework. Overall, the findings indicate that the DT-SF TOPSIS model provides a practical, confidence-weighted, and uncertainty-aware tool for multi-criteria diabetes risk assessment, with possible applications to other chronic diseases.