The Role of Artificial Intelligence in General Surgery: A Systematic Review and Meta-Analysis of Machine Learning Applications in Colorectal Cancer Treatment Outcomes

Abstract

Colorectal cancer (CRC) is a leading cause of global cancer morbidity and mortality, with surgical resection as the primary curative treatment. The integration of artificial intelligence (AI), particularly machine learning (ML), into CRC surgery presents a promising avenue for improving patient care through enhanced prediction and precision. This systematic review and meta-analysis aimed to synthesize evidence on the impact of ML on CRC surgical outcomes. A comprehensive search of PubMed, Scopus, Embase, Cochrane Library, and Web of Science was conducted on September 1, 2025, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Ultimately, 10 studies were included in the qualitative synthesis, with a subset used in the meta-analysis. The results indicated that ML-assisted or robotic surgeries were associated with a nonsignificant reduction in postoperative complications (risk ratio (RR) 0.85, 95% CI 0.70-1.03) compared to conventional methods. However, robotic procedures were significantly linked to longer operative times (mean difference (MD) +45.2 min, 95% CI 28.5-61.9). The meta-analysis of predictive model performance yielded a pooled area under the curve (AUC) of 0.84 (95% CI: 0.80-0.88), suggesting good overall discriminatory ability, though this finding is tempered by substantial heterogeneity (I²=68%) and a high risk of bias across all included studies. In conclusion, while ML applications in CRC surgery show potential, current evidence does not confirm significant superiority in reducing complications. The increased operative time and methodological limitations of existing research highlight the need for more rigorous, high-quality trials to fully ascertain the clinical value of these technologies.

Ähnliche Arbeiten