Integration of Scheimpflug-based corneal tomography and biomechanical assessments for enhancing ectasia detection
Session Details
Session Title: Moderated Poster Session: New and Interesting
Venue: Poster Village: Pod 1
First Author: : R.Ambrosio BRAZIL
Co Author(s): : B. Lopes J. Buhren C. Roberts A. Elsheikh R. Vinciguerra P. Vinciguerra
Abstract Details
Purpose:
To present the Tomographic/Biomechanical Index (TBI), that combines Scheimpflug-based corneal tomography and biomechanics for enhancing ectasia detection.
Setting:
1. Rio de Janeiro Corneal Tomography and Biomechanics Study Group
2. Federal University of São Paulo
3. Eye Center, Humanitas Clinical and Research Center, Via Manzoni 56, Rozzano (MI); Italy
Methods:
Patients from different continents were studied. One eye randomly selected from 480 patients with normal corneas and from 204 keratoconus patients comprised groups I and II respectively. Group III included 72 non-operated ectatic eyes from 94 patients with very asymmetric ectasia, whose fellow eyes (group IV) presented with normal topography. Pentacam HR and Corvis ST (OCULUS; Wetzlar, Germany) parameters were analyzed and combined using different artificial intelligence methods (AI). The accuracies for detecting ectasia of BAD-D (Belin/Ambrósio Deviation) and CBI (Corvis Biomechanical Index) were compared to TBI, considering the areas under receiver operating characteristic curves (AUROC).
Results:
The random forest method with leave-one-out cross-validation (RF/LOOCV) provided the best AI model. The AUROC for detecting ectasia (groups II, III and IV) of TBI was 0.996, being statistically higher (DeLong, p<0.001) than BAD-D (0.956) and CBI (0.936). TBI cutoff value of 0.79 provided 100% sensitivity for detecting clinical ectasia (groups II and III) with 100% specificity. Considering group IV, AUROC for TBI, BAD-D and CBI were 0.985, 0.839 and 0.822 (DeLong, p<0.001). An optimized TBI cutoff value of 0.29 provided 90.4% sensitivity in group IV, with 96% specificity.
Conclusions:
TBI generated by RF/LOOCV provides accuracy for detecting ectasia, exceeding other techniques. TBI is sensitive for detecting sub-clinical (fruste) ectasia among eyes with normal topography in very asymmetric patients. TBI may also confirm unilateral disease, potentially characterizing the inherent ectasia susceptibility of the cornea.
Financial Disclosure:
gains financially from product or procedure presented, is employed by a for-profit company with an interest in the subject of the presentation
