Related Experiment Video
Updated: May 8, 2026
07:23
Optimization of the Retinal Vein Occlusion Mouse Model to Limit Variability
Published on: August 6, 2021
Artificial Intelligence-Based Prognostic Models for Postoperative Outcomes in Vitreoretinal Surgery: A Systematic
Abdullah Al-Ani1, Liam Connors2, David Mikhail3
1Section of Ophthalmology, Department of Surgery, Cumming School of Medicine, University of Calgary, Calgary, Canada.
Ophthalmology. Retina
|May 7, 2026
Summary
Artificial intelligence (AI) models show promise for predicting vitreoretinal surgery outcomes, outperforming traditional methods. Further validation is needed to confirm their clinical utility in surgical planning and patient counseling.
Area of Science:
- Ophthalmology
- Medical Artificial Intelligence
- Surgical Outcomes Prediction
Background:
- Vitreoretinal surgery outcomes prediction is crucial for patient management.
- Conventional statistical methods have limitations in analyzing complex data.
- Artificial intelligence (AI) offers potential for improved predictive accuracy.
Purpose of the Study:
- To evaluate the performance of AI models in predicting outcomes after vitreoretinal surgery.
- To compare AI model performance against conventional statistical approaches.
- To assess the potential of AI in augmenting preoperative prognostication.
Main Methods:
- A systematic review and meta-analysis of studies predicting vitreoretinal surgical outcomes using AI.
- Searches conducted across multiple databases (MEDLINE, Embase, Cochrane, etc.) and grey literature.
- Risk of bias assessed using PROBAST and evidence quality using GRADE.
Main Results:
- 26 studies (18,724 eyes) met eligibility criteria; 12 included in meta-analysis.
- AI models, particularly deep learning (e.g., CNNs), often outperformed conventional methods.
- Meta-analysis showed pooled sensitivity of 0.89 and specificity of 0.87 for AI models, with low certainty evidence.
Conclusions:
- AI-based models demonstrate potential for forecasting vitreoretinal surgical outcomes.
- AI can aid in clinical decision-making, surgical planning, and patient counseling.
- Further external validation and clinical implementation studies are necessary.