Research Progress of MRI-based Radiomics in Rectal Cancer

Area of Science:

• Oncology and Radiology
• Medical Imaging Analysis
• Artificial Intelligence in Medicine

Background:

• Rectal cancer (RC) is a prevalent malignancy with significant global mortality.
• Radiomics transforms medical images into mineable data using algorithms like filter-based methods and texture analysis.
• Machine learning and deep learning integrate radiomic features for objective diagnostic and prognostic insights.

Purpose of the Study:

• To provide a comprehensive overview of current MRI radiomics applications in rectal cancer.
• To compare feature extraction methods used in MRI radiomics for rectal cancer.
• To critically discuss existing challenges and future research directions in the field.

Main Methods:

• Utilizing multi-parametric magnetic resonance imaging (MRI) for detailed rectal cancer characterization.
• Employing high-throughput extraction algorithms for radiomic feature generation.
• Integrating radiomic features with machine learning or deep learning models.

Main Results:

• MRI-based radiomics shows promise in tumor-node-metastasis (TNM) staging, risk factor identification, and genetic marker prediction.
• Applications extend to evaluating neoadjuvant therapy response and prognostic survival analysis in rectal cancer.
• The review compares various feature extraction techniques and discusses their efficacy.

Conclusions:

• MRI radiomics is a rapidly advancing field with substantial potential for enhancing rectal cancer management.
• Further research is needed to address current limitations and explore novel developments in MRI radiomics for rectal cancer.
• Standardization of feature extraction and validation across diverse datasets are crucial for clinical translation.