What does this research mean for the field?

Machine learning models using clinico-radiological variables predict pathological tumor invasion and nodal status in rectal cancer with performance similar to radiologists, and the addition of MRI-based radiomics does not significantly improve this predictive performance. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To evaluate the performance of machine learning models combining clinical-radiological and radiomic MRI data for predicting pathological tumor invasion and nodal status in rectal cancer.

September 8, 2026Applied SciencesOpen Access

MRI-Based Radiomics and Machine Learning for Predicting Pathological Tumor Invasion and Nodal Status in Rectal Cancer: A Retrospective Study

View Full Paper

Ask AI

Bookmark

Authors

MCM. Cerezo DCDavid López Cornejo AOAlba Ortigosa‐Palomo

Discussion

Member takes

Overview

Retrospective study finds comparable tumor and nodal staging accuracy between radiomics and standard models in rectal cancer, suggesting radiomics adds limited diagnostic value.

Key Points

To evaluate the performance of machine learning models combining clinical-radiological and radiomic MRI data for predicting pathological tumor invasion and nodal status in rectal cancer.
Retrospective observational study of 152 rectal cancer patients, including 70 without neoadjuvant therapy and 82 with neoadjuvant therapy.
Extracted radiomic features from preoperative high-resolution T2-weighted MRI using two segmentation protocols: tumor alone and tumor plus mesorectum.
Evaluated 23 machine learning algorithms using cross-validation to select optimal predictive models based on area under the receiver operating characteristic curve (AUC).
Clinical and radiological models achieved the most consistent predictive performance in the overall cohort, with AUCs of 0.767 for tumor invasion and 0.764 for nodal status.
Radiomic and combined models showed moderate and variable performance, reaching peak AUCs of 0.770 for tumor invasion and 0.732 for nodal status without demonstrating significant improvement over clinical evaluation.

Cite This Study

Cerezo et al. (2026) studied this question.

synapsesocial.com/papers/6a9fd7d558e84d0ff5b46eb4 https://doi.org/10.3390/app16178897

View Full Paper

Ask AI

Bookmark

September 8, 2026Applied SciencesOpen Access

MRI-Based Radiomics and Machine Learning for Predicting Pathological Tumor Invasion and Nodal Status in Rectal Cancer: A Retrospective Study

View Full Paper

Ask AI

Bookmark

Authors

MCM. Cerezo DCDavid López Cornejo AOAlba Ortigosa‐Palomo

Discussion

Member takes

Overview

Retrospective study finds comparable tumor and nodal staging accuracy between radiomics and standard models in rectal cancer, suggesting radiomics adds limited diagnostic value.

Key Points

To evaluate the performance of machine learning models combining clinical-radiological and radiomic MRI data for predicting pathological tumor invasion and nodal status in rectal cancer.
Retrospective observational study of 152 rectal cancer patients, including 70 without neoadjuvant therapy and 82 with neoadjuvant therapy.
Extracted radiomic features from preoperative high-resolution T2-weighted MRI using two segmentation protocols: tumor alone and tumor plus mesorectum.
Evaluated 23 machine learning algorithms using cross-validation to select optimal predictive models based on area under the receiver operating characteristic curve (AUC).
Clinical and radiological models achieved the most consistent predictive performance in the overall cohort, with AUCs of 0.767 for tumor invasion and 0.764 for nodal status.
Radiomic and combined models showed moderate and variable performance, reaching peak AUCs of 0.770 for tumor invasion and 0.732 for nodal status without demonstrating significant improvement over clinical evaluation.

Cite This Study

Cerezo et al. (2026) studied this question.

synapsesocial.com/papers/6a9fd7d558e84d0ff5b46eb4 https://doi.org/10.3390/app16178897

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark