Abstract
Radiomics involves the extraction of quantitative information from medical images and can offer valuable insights into the management of rectal cancer (RC). We aim to systematically review the current literature on radiomics' role in the prediction of lymph nodes and distance metastases, local recurrence and survival outcomes in rectal cancer. Method: This systematic review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and principles. We systematically searched PubMed, Cochrane, Google Scholar, and ResearchGate databases to identify the relevant articles. Result: After searching 3327 articles, only 50 articles were included after assessing the quality using a radiomics quality score (QRS) and finding a mean score of 14.74. Only eight studies used CT radiomics, with the rest based on MRI radiomics. Only three of the included studies are prospective, with the others being retrospective cohort studies and 14 having external validation. Most of the included studies concluded that radiomic models alone or combined models achieve better outcome predictions when compared with clinical and subjective analysis. Conclusions: Radiomics offers significant promise as a non-invasive tool for predicting lymph node status, distant metastasis (DM), recurrence, and survival outcomes in rectal cancer. There is a need for more robust prospective studies with cost benefit analysis for implementation into clinical practice.