Objective: To investigate the diagnostic and predictive value of MRI features combined with clinical indicators for prostate cancer (PCa) and clinically significant prostate cancer (csPCa), and to establish a non-invasive combined model. Methods: A total of 36 patients with pathologically confirmed benign lesions (44 foci) and 23 patients with PCa (49 foci), including 25 foci of csPCa and 68 foci of non-csPCa, were included. SyMRI quantitative maps and clinical indicators were collected, and 224 imaging features were extracted. The intra- and inter-group correlation coefficients (ICC) for each feature were calculated using intra- and inter-group correlation analysis, and features with an ICC > 0.75 were selected as stable features that could be reproducibly extracted. Independent predictors were screened using logistic regression to construct single and combined models, and the performance was evaluated using ROC curves. Results: Age, PSAD, PD map contrast, and T2 map joint entropy were significantly higher in the PCa group compared to the benign group, while the median ADC was significantly lower (p < 0.05). The above-mentioned indicators were significantly correlated with PCa and csPCa, and the diagnostic performance of the combined model was superior to that of a single MRI or clinical model. Conclusion: MRI features combined with PSAD can effectively differentiate PCa and predict csPCa, providing a non-invasive quantitative diagnostic basis for clinical practice.