Loading
Radiomics and Machine Learning in Medical Imaging / Colorectal Cancer Surgical Treatments / Colorectal and Anal Carcinomas · Journal article
Applied Sciences · September 7, 2026
Early or partial results. Treat as a signal, not a conclusion.
This retrospective study applied 23 machine learning algorithms to MRI-based clinical, radiological, and radiomic features in 152 rectal cancer patients to predict pathological tumor invasion (pT) and nodal status (pN). Clinical and radiological variables alone achieved AUCs of 0.767 for pT and 0.764 for pN, while radiomic and combined models showed no meaningful improvement, suggesting that radiomics does not add predictive value over conventional imaging interpretation in this setting.
Retrospective observational cohort study with machine learning model development and cross-validation. Patients with rectal cancer undergoing preoperative MRI for local staging; stratified by neoadjuvant therapy receipt (yes/no).. Intervention: Machine learning models trained on clinical, radiological, and radiomic features extracted from preoperative T2-weighted MRI. Compared with: No explicit comparator cohort; models compared internal performance to each other (clinical-radiological vs. radiomic vs. combined). n = 152.
Clinical and radiological variables alone achieved AUC 0.767 for pT prediction in overall cohort Clinical and radiological variables achieved AUC 0.764 for pN prediction in overall cohort Radiomic models achieved maximum AUC 0.770 for pT (minimal improvement over clinico-radiological model)
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
Clinicians should recognize that standard clinical and radiological variables analyzed with machine learning perform comparably to radiomic analysis for predicting pT and pN in rectal cancer; radiomics does not appear to offer additional predictive value in this retrospective cohort and should not yet replace conventional staging interpretation. External prospective validation would be needed before any clinical implementation.
A retrospective single-center study using cross-validation in a small, heterogeneous cohort (152 patients, split by treatment) that reports surrogate endpoints (AUC) rather than clinical outcomes; radiomic findings did not improve on baseline models, limiting clinical utility.
As stated by the source record.
Quoted from the source exactly as published.
Clinicians should recognize that standard clinical and radiological variables analyzed with machine learning perform comparably to radiomic analysis for predicting pT and pN in rectal cancer; radiomics does not appear to offer additional predictive value in this retrospective cohort and should not yet replace conventional staging interpretation. External prospective validation would be needed before any clinical implementation.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
Introduction: MRI is the gold-standard imaging modality for rectal cancer (RC) local staging, but the ability to determine tumor invasion (pT) and nodal status (pN) remains limited in clinical practice. The main objective of this study is to evaluate the performance of different machine learning (ML) models based on clinical–radiological and radiomic variables for predicting these categories from preoperative MRI. Methods: A retrospective observational study was conducted involving 152 patients with RC (70 without neoadjuvant therapy and 82 with neoadjuvant therapy). Radiomic features were extracted from high-resolution T2 sequences using two independent segmentations: tumor and tumor + mesorectum. Twenty-three ML algorithms were evaluated using cross-validation to predict pT and pN. For each combination of outcome, cohort, data source and segmentation, an optimal model was selected based on the area under the curve (AUC). Results: Models based on clinical and radiological variables showed the most consistent performance, particularly in the overall cohort, with AUCs of 0.767 for pT and 0.764 for pN. The radiomic and combined models achieved a moderate and heterogeneous performance, with maximum AUCs of 0.770 for pT and 0.732 for pN. Conclusions: The clinico-radiological variables analyzed using ML showed a predictive performance similar to that of a radiologist. Radiomics did not show significant improvement in this setting.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.