performance analysis of mamdani and takagi-sugeno fuzzy systems for estimating the systolic blood pressure gradient in women using persian medicine pulsology

The systolic blood pressure gradient (sbpg) has shown significant diagnostic and prognostic value in assessing overall cardiac function. persian medicine pulsology (pm) offers a time-tested diagnostic framework; however, its integration with modern fuzzy-based sbpg analysis lacks systematic performance comparisons between inference architectures. objective: this study presents a comparative evaluation of the performance of mamdani and takagi-sugeno (ts) fuzzy inference systems for predicting systolic blood pressure gradient, considering diagnostic parameters from persian medicine (pm) pulsology. methods: both architectures were implemented in matlab r2021b and validated using clinical pm pulse data, including speed, frequency, strength, and length as input variables and sbpg as the output variable. mamdani and ts fuzzy systems were used, and their performance was evaluated using mae, rmse, and mre criteria. key findings: a comparative analysis revealed a significant accuracy advantage for the ts system (mae = 0.1838, rmse = 0.2495 ,mre = 2.96%) over the mamdani system (mae = 0.2700, rmse = 0.3337, mre = 4.11%).innovative contributions: the study s primary contributions are threefold: (1) pioneering the use of pm pulsology in fuzzy-based systolic bp gradient forecasting, (2) establishing a clinically-validated and structurally stable framework for both models, and (3) open-sourcing the complete mamdani and ts model implementations.