Related Experiment Video
Updated: Jun 19, 2026
14:56
An Experimental Paradigm for the Prediction of Post-Operative Pain PPOP
Published on: January 27, 2010
21.3K
Predicting Surgical Difficulty in Rectal Cancer Surgery: A Systematic Review of Artificial Intelligence Models
Conor Hardacre1,2, Thomas Hibbs1, Matthew Fok1,2
1University Hospitals of Liverpool Group, Liverpool L7 8YE, UK.
Cancers
|March 13, 2025
Summary
Artificial intelligence (AI) analyzing preoperative MRI scans can predict rectal cancer surgery difficulty. These AI tools show strong performance and warrant further clinical evaluation for personalized surgical approaches.
Area of Science:
- Oncology
- Radiology
- Surgical Technology
Background:
- Minimally invasive surgery offers various rectal cancer resection modalities, with robotic surgery being the most advanced.
- Surgical approach decisions currently rely on local resources and surgeon expertise, limiting access to advanced techniques.
- Predictive tools for surgical difficulty using preoperative data are needed to optimize resource allocation, especially for robotic surgery.
Purpose of the Study:
- To systematically review and appraise existing literature on artificial intelligence (AI)-driven preoperative MRI analysis for predicting rectal cancer surgery difficulty.
- To identify knowledge gaps and promising AI models for further clinical evaluation.
- To support efficient resource utilization and personalized surgical strategies.
Main Methods:
- A systematic review and narrative synthesis adhering to PRISMA and SWiM guidelines.
- Searches conducted on Medline, Embase, and CENTRAL Trials register for studies from 2012-2024.
- Included studies utilized AI on preoperative MRI of adult rectal cancer patients to stratify surgical difficulty; data extracted on study characteristics, AI design, and performance metrics.
Main Results:
- 40 studies were included from 568 initial articles.
- AI models were identified for eight domains: surgical difficulty grading, extramural vascular invasion (EMVI), lymph node metastasis (LNM), lymphovascular invasion (LVI), perineural invasion (PNI), T staging, and stapler firings.
- Multiple models demonstrated very good (AUC >0.80) to excellent performance in predicting these surgical difficulty parameters.
Conclusions:
- AI tools analyzing preoperative rectal MRI are emerging as valuable aids for assessing surgical difficulty in rectal cancer.
- The progressing development and strong performance of these AI models indicate significant potential.
- Further clinical evaluation is recommended to integrate these tools into personalized surgical planning and optimize resource management.
