Related Experiment Video
Updated: Jun 2, 2025
07:12
Development of a Gaze-Contingent Display Framework Designed for Perceptual and Oculomotor Research with Simulated Central Vision Loss
Published on: April 11, 2025
294
Supervised machine learning statistical models for visual outcome prediction in macular hole surgery: a
Kanika Godani1, Vishma Prabhu1, Priyanka Gandhi1
1Department of Retina and Vitreous, Narayana Nethralaya, #121/C, 1st R Block, Chord Road, Rajaji Nagar, Bengaluru, 560010, India.
International Journal of Retina and Vitreous
|January 13, 2025
Summary
Random Forest regression accurately predicts visual acuity after macular hole surgery using optical coherence tomography data. This machine learning approach aids surgical planning and patient counseling for better outcomes.
Area of Science:
- Ophthalmology
- Medical Imaging
- Machine Learning
Background:
- Macular hole (MH) surgery aims to restore visual acuity (VA).
- Preoperative optical coherence tomography (OCT) provides key structural data for MHs.
- Predicting postoperative VA is crucial for surgical planning and patient expectations.
Purpose of the Study:
- To assess the predictive accuracy of various machine learning (ML) models for postoperative VA following MH surgery.
- To utilize preoperative OCT parameters for forecasting visual outcomes.
Main Methods:
- Retrospective analysis of 158 eyes with full-thickness MHs.
- Extraction of OCT-derived qualitative and quantitative MH characteristics.
- Training and testing six supervised ML models (including Random Forest regression) on 14,652 OCT data points.
Main Results:
- 91% of patients achieved MH closure with a median VA gain of -0.3 logMAR.
- Random Forest (RF) regression model showed the highest predictive accuracy (lowest MSE).
- Postoperative MH closure status and MH area index were the most significant predictors of VA.
Conclusions:
- The RF regression model offers superior predictive accuracy for postoperative VA.
- ML-driven predictions using preoperative OCT data can enhance surgical planning and patient counseling.
- This approach provides reliable insights into expected visual outcomes after MH surgery.
