AUTHOR=Toth Bertalan , Mesterházi Marcell , Szabo Istvan , Mersich Tamas , Toth Erika , Szoke Janos , Benczur Andras , Kerepesi Csaba , Tarnoki David Laszlo , Tarnoki Adam Domonkos 
  
TITLE=Magnetic resonance imaging based radiomics for predicting pathogenetic features and survival in rectal cancer
  
JOURNAL=Pathology and Oncology Research
  
VOLUME=Volume 32 - 2026
  
YEAR=2026
  
URL=https://www.por-journal.com/journals/pathology-and-oncology-research/articles/10.3389/pore.2026.1612303
  
DOI=10.3389/pore.2026.1612303
  
ISSN=1532-2807
  
ABSTRACT=BackgroundPrediction of pathologic features and outcome in patients with rectal cancer is challenging as a result of lack of a significant biomarker and heterogeneity between and within tumors. This study aims to evaluate the potential of Magnetic Resonance Imaging (MRI)-based radiomics in predicting key pathological features and long-term survival outcomes in patients.MethodsA retrospective study was conducted on 510 rectal cancer patients treated between 2015 and 2019. The inclusion criteria required pre-therapeutic MRI performed on a Discovery MR750W 3.0T machine and known KRAS mutation status. Forty-seven patients met the criteria. MRI sequences included T1-weighted, T2-weighted fat-saturated (T2FS), high-resolution T2-weighted (T2HR), and diffusion-weighted imaging (DWI). Radiomic features were extracted using PyRadiomics, and machine learning models were developed using XGBoost and LightGBM classifiers. Feature selection was performed using Sequential Feature Selector (SFS) and Minimum Redundancy Feature Selection (mRMR).ResultsThe model for KRAS mutation status achieved an Area Under the ROC curve (AUC) of 0.7475 (training) and 0.75 (testing). Lymph node invasion prediction had an AUC of 0.7892 (training) and 0.7984 (testing). Vascular invasion prediction yielded an AUC of 0.6989 (training) and 0.7143 (testing). The 5-year survival prediction model showed an AUC of 0.7848 (training) and 0.7750 (testing). Metastasis prediction achieved an AUC of 0.6627 (training) and 0.6857 (testing).ConclusionMRI-based radiomics demonstrates significant potential in predicting key pathological features and long-term survival outcomes in rectal cancer patients. Integrating multimodal imaging data and clinical information, along with automated segmentation techniques, could further enhance model accuracy and clinical utility.