What question did this study set out to answer?

The study aims to develop a deep learning model that predicts pathological complete response (pCR) in rectal cancer after treatment.

Automated deep learning model for predicting pathological complete response in rectal cancer: A tool to organ-preserving strategies

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates high-sensitivity prediction of pCR in rectal cancer, suggesting utility in assessment strategies.

The study aims to develop a deep learning model that predicts pathological complete response (pCR) in rectal cancer after treatment.
Developed a segmentation-free deep learning model using a 3D siamese convolutional neural network.
Analyzed paired pre- and post-nCRT T2-weighted MRI volumes from 195 patients, divided into training and test sets.
Evaluated performance using AUC-ROC and classification metrics in an independent test set of 49 patients.
Achieved an AUC-ROC of 0.71 (95% CI: 0.55-0.85) for pCR prediction in the test set.
Sensitivity was 100% (95% CI: 70.1-100) and negative predictive value was 100% (95% CI: 81.6-100) at the selected threshold.
Specificity was 42.5% (95% CI: 28.5-57.8), indicating variability based on pCR prevalence in the cohort.

J. et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark