 may offer a more accurate method for detecting the spread of prostate cancer beyond the prostate capsule){kind=link}
A study suggests that artificial intelligence (AI) may offer a more accurate method for detecting the spread of prostate cancer beyond the prostate capsule, a condition known as extracapsular extension (ECE), compared to traditional imaging methods and prediction tools.
Researchers analyzed data from 147 patients who underwent MRI-targeted biopsy and prostate removal surgery between 2016 and 2022. Using AI-generated cancer maps (Unfold AI by Avenda Health), they compared its ability to predict ECE with conventional methods like MRI scans and established prediction tools, such as capsular contact length measurements and the “PRedicting ExtraCapsular Extension” nomogram.
The study found that Unfold AI had a significantly higher accuracy in predicting ECE, particularly in the posterior part of the prostate, where it outperformed MRI. The AI system had an area under the curve score of 0.81, which was better than other methods for patient-level ECE detection. It achieved 68% sensitivity (correctly identifying cancer spread when it was present) and 78% specificity (correctly ruling out cancer spread when it was absent). The AI-based cancer map also had fewer false negatives, especially in harder-to-detect areas of the prostate.
By offering more precise cancer staging, AI tools like Unfold AI could help doctors better plan prostate cancer surgeries, reducing the chances of nerve damage and improving patient outcomes.