Researchers developed machine learning models that outperform PSA testing in predicting abnormal prostate MRI findings for suspected prostate cancer.
Key Details
- 1Study involved 11,879 prostate MRI scans for suspected prostate cancer.
- 2Two ML models were trained, with AUCs of 0.711 (Model A) and 0.616 (Model B), compared to 0.593 for PSA testing alone.
- 3Model A included prostate volume in addition to clinical variables; Model B did not.
- 4Model A had higher specificity (28.3%) and comparable sensitivity (89%) versus PSA testing (>4 ng/mL).
- 5False negative rates: 8% for Model A, 16.8% for Model B; most were clinically insignificant or benign.
Why It Matters
By improving risk stratification with prostate MRI, machine learning models can help triage patients more effectively, optimizing resource use and potentially reducing unnecessary imaging. This supports more personalized and efficient approaches in prostate cancer diagnosis within radiology.
Source
AuntMinnie
Related News
•AuntMinnie
Deep Learning AI Outperforms Radiologists in Detecting ENE on CT
A deep learning tool, DeepENE, exceeded radiologist performance in identifying lymph node extranodal extension in head and neck cancers using preoperative CT scans.
•Radiology Business
Patients Favor AI in Imaging Diagnostics, Hesitate on Triage Use
Survey finds most patients support AI in diagnostic imaging but are reluctant about its use in triage decisions.
•AuntMinnie
AI Projected to Reshape Radiologist Workload But Not Eliminate Jobs
Stanford researchers predict AI could reduce radiologist hours by up to 49% over the next five years, though workforce size is likely to remain stable due to rising imaging demand.