Diagnosis of Coronary Heart Disease Through Fuzzy Information Measures and Pattern Recognition-Based Segmentation and Localization in Computed Tomography Angiography

Timely detection of coronary heart disease (CHD) is both significant and challenging. Advanced computational techniques such as similarity measures (SMs), entropy measures, and distance-measure-based segmentation can greatly enhance the precision, speed, and reliability of computed tomography angiography (CTA) image interpretation, thereby supporting more effective early detection and management of CHD. The concept of the interval-valued spherical fuzzy set (Iv-SFS) serves as a powerful tool for handling complex and ambiguous information, as it allows the representation of membership, abstinence, and non-membership values in the form of intervals. Building on the structure of Iv-SFS, this study introduces novel SMs, including interval-valued spherical fuzzy cosine SMs (Iv-SFCSM), Iv-SF cosine weighted SMs (Iv-SFCWSM), Iv-SF dice SMs (Iv-SFDSM), and Iv-SF dice weighted SMs (Iv-SFDWSM). Several axioms of SMs are demonstrated to verify the validity of the proposed measures. The developed SMs are then applied to solve a multi-attribute decision-making (MADM) problem for CHD diagnosis. Key features of the proposed approach are discussed, and comparative analyses with existing SMs highlight its advantages. The paper concludes with remarks underscoring the effectiveness and applicability of the proposed framework.