Development and Validation of the Relational Tissue Altered (RTA) Index: Applied Artificial Intelligence for the...

Area of Science:

• Ophthalmology
• Biomedical Engineering
• Artificial Intelligence

Background:

• A retrospective case-control study analyzed data from 3278 stable eyes and 105 eyes with postoperative ectasia.
• Investigated the impact of laser vision correction (LVC) on corneal structure.

Purpose of the Study:

• To develop a predictive computational model for assessing LVC's impact on corneal structure.
• To objectively and accurately represent the biomechanical changes post-refractive surgery.

Main Methods:

• Utilized an artificial intelligence-based machine learning approach.
• Employed knowledge discovery in databases (KDD) principles for model development.
• Evaluated model performance using receiver operating characteristic (ROC) curves, area under the curve (AUC), sensitivity, and specificity.

Main Results:

• The Relational Tissue Altered (RTA) index demonstrated superior performance with an AUC of 0.913.
• RTA showed higher AUC than Residual Stromal Bed (RSB) and Percent Tissue Altered (PTA) metrics.
• RTA achieved 76.0% sensitivity and 89.2% specificity.

Conclusions:

• The Relational Tissue Altered (RTA) index offers an objective, data-driven measure of LVC's structural impact on the cornea.
• RTA outperforms traditional parameters like RSB and PTA in characterizing surgical effects.
• Integrating RTA with preoperative assessments can enhance risk stratification and personalize refractive surgery.