Clinical-radiomics hybrid modeling outperforms conventional models: machine learning enhances stratification of adverse prognostic features in prostate cancer
This research demonstrates improved risk stratification of adverse features in prostate cancer, highlighting the role of machine learning and MRI analysis.
Key Points
The hybrid model combining radiomics and clinical features achieved the highest AUC of 0.909 in predicting adverse pathological features.
Random forest emerged as the best-performing algorithm, significantly outperforming conventional clinical models.
A cohort of 137 prostate cancer patients was analyzed, revealing the effectiveness of MRI-based machine learning in stratification.
Integration of clinical characteristics with radiomics offers a non-invasive method for personalized treatment planning.