Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study introduces TRACE-PCa, a novel end-to-end temporal and multimodal model that predicts pathological progression in prostate cancer during active surveillance by leveraging longitudinal MRI data without the need for explicit lesion segmentation. By encoding serial scans with a pretrained 3D MRI foundation model and employing a temporal attention gate, the model enhances the detection of significant imaging changes over time. Validated on a longitudinal cohort, TRACE-PCa outperforms existing risk-stratification tools and aligns closely with radiologist assessments, indicating its potential to minimize unnecessary biopsies while maintaining diagnostic accuracy.
TRACE-PCa can significantly reduce unnecessary biopsies in prostate cancer surveillance by accurately predicting disease progression from longitudinal MRI data.
Active surveillance (AS) is the preferred strategy for favorable-risk prostate cancer, yet current protocols rely on scheduled repeat biopsies, most of which reveal no progression and are unnecessary. Existing risk-stratification tools operate on single time-point imaging or depend on explicit lesion segmentation, limiting their ability to capture longitudinal change and excluding patients without an MRI-visible lesion. In this study, we propose an end-to-end temporal and multimodal model for predicting pathological progression during AS without lesion segmentation. We encode each serial scan with a pretrained 3D MRI foundation model and introduce a temporal attention gate that recalibrates the multi-visit features to amplify focal imaging changes associated with progression. The gated imaging representation is then fused with clinical variables in a multimodal framework to estimate the probability of progression. Validated on a longitudinal AS cohort, our approach consistently outperforms competing baselines and performs comparably to the radiologist assessment representing current clinical practice. It maintains high negative predictive value while achieving higher positive predictive value, demonstrating its potential to safely reduce unnecessary biopsies during surveillance.