Extracapsular extension risk assessment using an artificial intelligence prostate cancer mapping algorithm
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Artificial intelligence (AI) cancer mapping accurately predicted extracapsular extension (ECE) in prostate cancer (PCa), outperforming traditional MRI and nomograms. This AI tool shows promise for improving surgical planning and patient outcomes.
Area Of Science
- Urology
- Oncology
- Medical Imaging
Background
- Accurate detection of extracapsular extension (ECE) in prostate cancer (PCa) is crucial for treatment planning and predicting outcomes.
- Current methods including MRI and nomograms have limitations in precisely identifying ECE.
- AI-driven tools offer potential for enhanced diagnostic accuracy in oncology.
Purpose Of The Study
- To compare the detection rates of PCa ECE using AI-generated cancer maps versus conventional MRI and nomograms.
- To evaluate the performance of AI software in predicting ECE risk at both patient and quadrant levels.
- To assess the potential of AI in improving PCa staging and risk stratification.
Main Methods
- Retrospective analysis of 147 patients undergoing MRI-targeted biopsy and radical prostatectomy.
- Utilized FDA-cleared AI software (Unfold AI) for 3D cancer probability mapping and ECE risk estimation.
- Compared AI predictions against conventional ECE predictors (MRI Likert scores, capsular contact length, PSMA T stage, Partin tables, and a dedicated nomogram).
- Evaluated performance using receiver operator characteristic curves and DeLong's test.
Main Results
- Unfold AI demonstrated a significantly higher area under the curve (AUC=0.81) for patient-level ECE prediction compared to other methods.
- At the quadrant level, Unfold AI outperformed MRI Likert scores in posterior (AUC=0.89 vs 0.82) and all quadrants (AUC=0.89 vs 0.82).
- Unfold AI exhibited improved sensitivity and specificity, with a notably lower false negative rate than MRI, particularly in posterior quadrants.
Conclusions
- AI-generated cancer mapping (Unfold AI) accurately predicts ECE risk in PCa, surpassing conventional methods.
- The AI tool significantly enhances ECE detection over MRI, especially in posterior prostate regions.
- Improved PCa staging and risk stratification through AI mapping can potentially lead to better oncological and functional outcomes, including informed nerve-sparing surgery.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.