Ranking polycystic ovarian syndrome (PCOS) drugs using degree-based indices in QSPR models and CRITIC-driven MCDM methods.

Globally, 8-13% of reproductive-age women experience PCOS, a complicated endocrine condition. This study investigates several methods for treating polycystic ovary syndrome (PCOS) as well as its causes, which include a complicated interplay between hereditary susceptibility, hormonal imbalances, insulin resistance, and lifestyle variables. An advanced graph-theory technique is used in the investigation of PCOS medication chemical structure prediction, with special focus on unique degree-based topological indices like the Banhatti and Zagreb indices. The efficacy of fifteen drugs, such as metformin, letrozole, spirolactone, etc., is evaluated in this study using QSPR analysis. We rate these medications using three different multi-criteria decision-making (MCDM) algorithms: CRITIC, CoCoSo, and MABAC. According to the results of the CoCoSo and MABAC analyses, the medicine orlistat has the best chance of success, and the MCDM methods have improved the process of evaluating and ranking treatment choices. The correlation between indices and properties falls within the range of 0.8-0.9, which shows a strong positive correlation with the physicochemical characteristics of compounds. The study demonstrates how topological indexes might improve the process of finding novel drugs and creating individualised treatment regimens for PCOS patients.