Optimized Artificial Intelligence for Enhanced Ectasia Detection Using Scheimpflug-Based Corneal Tomography and Biomechanical Data.

PURPOSE: To optimize artificial intelligence (AI) algorithms to integrate Scheimpflug-based corneal tomography and biomechanics to enhance ectasia detection. DESIGN: Multicenter cross-sectional case-control retrospective study. METHODS: A total of 3886 unoperated eyes from 3412 patients had Pentacam and Corvis ST (Oculus Optikgeräte GmbH) examinations. The database included 1 eye randomly selected from 1680 normal patients (N) and from 1181 "bilateral" keratoconus (KC) patients, along with 551 normal topography eyes from patients with very asymmetric ectasia (VAE-NT), and their 474 unoperated ectatic (VAE-E) eyes. The current TBIv1 (tomographic-biomechanical index) was tested, and an optimized AI algorithm was developed for augmenting accuracy. RESULTS: The area under the receiver operating characteristic curve (AUC) of the TBIv1 for discriminating clinical ectasia (KC and VAE-E) was 0.999 (98.5% sensitivity; 98.6% specificity [cutoff: 0.5]), and for VAE-NT, 0.899 (76% sensitivity; 89.1% specificity [cutoff: 0.29]). A novel random forest algorithm (TBIv2), developed with 18 features in 156 trees using 10-fold cross-validation, had a significantly higher AUC (0.945; DeLong, P < .0001) for detecting VAE-NT (84.4% sensitivity and 90.1% specificity; cutoff: 0.43; DeLong, P < .0001) and a similar AUC for clinical ectasia (0.999; DeLong, P = .818; 98.7% sensitivity; 99.2% specificity [cutoff: 0.8]). Considering all cases, the TBIv2 had a higher AUC (0.985) than TBIv1 (0.974; DeLong, P < .0001). CONCLUSIONS: AI optimization to integrate Scheimpflug-based corneal tomography and biomechanical assessments augments accuracy for ectasia detection, characterizing ectasia susceptibility in the diverse VAE-NT group. Some patients with VAE may have true unilateral ectasia. Machine learning considering additional data, including epithelial thickness or other parameters from multimodal refractive imaging, will continuously enhance accuracy. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.

Corneal deformation amplitude analysis for keratoconus detection through compensation for intraocular pressure and integration with horizontal thickness profile.

BACKGROUND: The Corvis ST provides measurements of intraocular pressure (IOP) and a biomechanically-corrected IOP (bIOP). IOP influences corneal deflection amplitude (DA), which may affect the diagnosis of keratoconus. Compensating for IOP in DA values may improve the detection of keratoconus. METHODS: 195 healthy eyes and 136 eyes with keratoconus were included for developing different approaches to distinguish normal and keratoconic corneas using attribute selection and discriminant function. The IOP compensation is proposed by dividing the DA by the IOP values. The first approaches include DA compensated for either IOP or bIOP and other parameters from the deformation corneal response (DCR). Another approach integrated the horizontal corneal thickness profile (HCTP). The best classifiers developed were applied in a validation database of 156 healthy eyes and 87 eyes with keratoconus. Results were compared with the current Corvis Biomechanical Index (CBI). RESULTS: The best biomechanical approach used the DA values compensated by IOP (Approach 2) using a linear discriminant function and reached AUC 0.954, with a sensitivity of 88.2% and a specificity of 97.4%. When thickness horizontal profile data was integrated (Approach 4), the best function was the diagquadratic, resulting in an AUC of 0.960, with a sensitivity of 89.7% and a specificity of 96.4%. There was no significant difference in the results between approaches 2 and 4 with the CBI in the training and validation databases. CONCLUSIONS: By compensating for the IOP, and with the horizontal thickness profile included or excluded, it was possible to generate a classifier based only on biomechanical information with a similar result to the CBI.