Model of a Support Vector Machine to Assess the Functional Cure for Surgery of Intermittent Exotropia

Area of Science:

• Ophthalmology
• Biomedical Engineering
• Data Science

Background:

• Basic intermittent exotropia requires timely intervention for functional cure.
• Assessing postoperative outcomes and identifying predictive factors is essential for optimizing treatment strategies.

Purpose of the Study:

• To explore the optimum timing for postoperative functional cure of basic intermittent exotropia using Support Vector Machine (SVM).
• To analyze factors influencing postoperative alignment and stereopsis restoration.

Main Methods:

• Prospective cross-sectional study with 132 patients and 6 months follow-up.
• Evaluations included angle of deviation, fusion, controllability, and stereoacuity.
• Statistical analyses involved chi-squared tests and logistic regression; SVM was used for prediction.

Main Results:

• The overall success rate for angle of deviation at 6 months was 63.6%.
• Postoperative day 1 angle of deviation was the sole significant factor for alignment at 6 months.
• Age of onset and preoperative distance stereoacuity influenced postoperative distance stereopsis restoration.

Conclusions:

• Postoperative day 1 angle of deviation is a critical predictor of alignment success.
• Support Vector Machine (SVM) model demonstrated 82.1% accuracy in predicting outcomes.
• The SVM model can aid in determining the optimal timing for functional cure in intermittent exotropia.