Deep learning models applied to PSMA-PET scans outperform existing tools in predicting two-year survival for metastatic prostate cancer.
Key Details
- 1Retrospective analysis of 486 PSMA-PET scans for metastatic prostate cancer patients from 2017-2025.
- 2Two models developed: an imaging-only model and a fusion model combining imaging and clinical variables.
- 3Fusion model achieved an AUC of 0.86, imaging model 0.82; both outperformed six established risk tools (AUC 0.43 to 0.65).
- 4Fusion model identified survival subgroups: 93% versus 43% two-year survival.
- 5Performance was consistent across radiotracers, treatment history, and metastatic burden.
- 6Researchers suggest prospective validation could enable personalized treatment selection.
Why It Matters
This demonstrates the potential of AI-driven PSMA-PET image analysis to enhance risk stratification and guide treatment for metastatic prostate cancer patients, surpassing conventional tools. Its adoption could personalize care and improve patient outcomes in clinical practice.

Source
AuntMinnie
Related News

•AuntMinnie
Deep Learning PET Biomarkers Outperform Clinical Tools in Prostate Cancer Survival Prediction
Deep learning-derived PET biomarkers more accurately predict metastatic prostate cancer survival than standard clinical risk models.

•AuntMinnie
Multinational AI Model Flags Osteoporosis on Chest X-rays
AI can opportunistically detect osteoporosis from chest x-rays across diverse international populations.

•AuntMinnie
LLM-Based Chart Summarization Tool Reduces Radiologist Review Time
An LLM-based chart summarization tool safely reduced radiologist study review time by 17%.