Development and validation of machine learning-based MRI radiomics models for preoperative lymph node staging in T3...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Accurate lymph node staging is crucial for rectal cancer (RC) treatment planning.
• Preoperative staging of lymph node metastasis in T3 RC remains challenging.
• Current methods may lack the precision needed for optimal therapeutic decisions.

Purpose of the Study:

• To evaluate the diagnostic performance of a magnetic resonance imaging (MRI)-based radiomics model for predicting lymph node staging in T3 RC patients.
• To assess the model's ability to differentiate between low (N0/N1) and high (N2) nodal burden.
• To determine the clinical utility of the radiomics model for preoperative staging.

Main Methods:

• Retrospective analysis of 225 T3 RC patients who underwent surgery without neoadjuvant therapy.
• Extraction of radiomics features from high-resolution T2-weighted imaging (T2WI) of primary tumors.
• Application of LASSO for feature selection and logistic regression for model construction, validated with ROC analysis and DCA.

Main Results:

• A radiomics signature using 16 selected features was developed.
• The logistic regression model achieved an AUC of 0.900 in the training cohort and 0.876 in the test cohort.
• Decision curve analysis confirmed the model's clinical utility for preoperative staging.

Conclusions:

• The developed MRI-based radiomics model offers an effective, noninvasive method for preoperative lymph node staging in T3 RC.
• The model accurately differentiates between N0/N1 and N2 stages, potentially improving treatment strategies.
• High-resolution T2WI radiomics shows promise for enhancing diagnostic accuracy in rectal cancer staging.