Clinical Report: Computer vision could turn surgical video into actionable clinical data
Overview
Research presented at Cleveland Clinic’s AI Summit indicates that computer vision can enhance the utility of surgical video by generating operative reports and aiding in the assessment of bladder lesions. AI-generated reports demonstrated higher accuracy compared to those written by surgeons.
Background
The integration of artificial intelligence in surgical practices is gaining traction, particularly in the realm of video analysis. This technology has the potential to improve surgical documentation and automate routine tasks.
Data Highlights
| Metric | AI-Generated Reports | Surgeon-Written Reports |
|---|---|---|
| Overall Accuracy | 87.3% | 72.8% |
| Meaningful Discrepancies | 12.7% | 27.2% |
Key Findings
- Computer vision models can identify surgical steps with 90% accuracy in robotic prostatectomy.
- AI-generated operative reports achieved 87.3% accuracy compared to 72.8% for surgeon-written reports (P = .001).
- AI can link operative reports directly to corresponding surgical video segments.
- A computer vision model predicted tumor histology with an AUC of 0.829 in the Mayo Clinic cohort.
- Automated instrument counting using computer vision can reduce delays at the end of surgical procedures.
Clinical Implications
The findings indicate that AI could enhance the accuracy of surgical documentation.
Conclusion
The exploration of computer vision in surgical contexts presents potential for enhancing surgical practice.
Related Resources & Content
- Abhinav Khana, Cleveland Clinic AI Summit, 2023 -- Computer vision could turn surgical video into actionable clinical data
- Advancements in Digital Operating Rooms: The Role of Video Technology in Surgical Procedures
- npj Digital Medicine — Extensive Self-Supervised Video Foundation Model for Enhanced Intelligent Surgical Procedures
- npj Digital Medicine — Computer vision applications in vascular surgery: a systematic review and critical appraisal
- Int. Journal of Computer Assisted Radiology and Surgery — SurgAnt-ViVQA: Utilizing GRU-Based Temporal Cross-Attention to Predict Surgical Events
- Advancements in Digital Operating Rooms: The Role of Video Technology in Surgical Procedures
- Extensive Self-Supervised Video Foundation Model for Enhanced Intelligent Surgical Procedures
- Computer vision applications in vascular surgery: a systematic review and critical appraisal
- SurgAnt-ViVQA: Utilizing GRU-Based Temporal Cross-Attention to Predict Surgical Events
- Multicentre randomized controlled trial exploring the clinical usefulness of the intraoperative use of an artificial intelligence–based anatomical navigation system
- Guidance on AI-enhanced surgical practice: a Delphi consensus on ontology, data, implementation and evaluation
- Guidances with Digital Health Content | FDA
This content is an AI-generated, fully rewritten summary based on a published scholarly article. It does not reproduce the original text and is not a substitute for the original publication. Readers are encouraged to consult the source for full context, data, and methodology.
