ESCRS - PO488 - Combining Spectral-Domain Oct And Air-Puff Tonometry Analysis To Diagnose Keratoconus Using Artificial Intelligence

Combining Spectral-Domain Oct And Air-Puff Tonometry Analysis To Diagnose Keratoconus Using Artificial Intelligence

Published 2022 - 40th Congress of the ESCRS

Reference: PO488 | Type: Free paper | DOI: 10.82333/v5vs-qk66

Authors: Nanji Lu* 1 , Ahmed Elsheikh 2 , Jos Rozema 3 , Ioannis Aslanides 4 , Carina Koppen 3 , Farhad Hafezi 5

1Wenzhou Medical University,Chengdu,China, 2University of Liverpool,Liverpool,United Kingdom, 3University of Antwerp,Antwerp,Belgium, 4Emmetropia Mediterranean Eye Institute, Heraklion, Greece,Herakloin,Greece, 5ELZA Institute,Zurich,Switzerland

Purpose

To investigate the diagnostic capacity of SD-OCT combined with air-puff tonometry using artificial intelligence (AI) in differentiating between normal and KC eyes.

Setting

Wenzhou Medical University, Wenzhou, Zhejiang, China

Methods

Patients who had either: undergone uneventful LVC with at least 3 years of stable follow-up, forme fruste keratoconus (FFKC), early keratoconus (EKC), or advanced keratoconus (AKC) were included. SD-OCT and biomechanical information from air-puff tonometry was divided into training and validation sets. AI models based on random forest (RF) or neural networks (NN) were trained to distinguish FFKC from normal eyes. Model accuracy was independently tested in FFKC and normal eyes. Receiver operating characteristic (ROC) curves were generated to determine area under the curve (AUC), sensitivity, and specificity values.

Results

223 normal eyes from 223 patients, 69 FFKC eyes from 69 patients, 72 EKC eyes from 72 patients, and 258 AKC eyes from 258 patients were included. The top AUC ROC values (normal eyes compared with AKC and EKC) were Pentacam Random Forest Index (PRFI) (AUC=0.985 and 0.958), Tomographic and Biomechanical Index (TBI) (AUC=0.983 and 0.925), and Belin-Ambrósio Deviation Index (BAD-D) (AUC=0.981 and 0.922). When SD-OCT and air-puff tonometry data were combined, the RF AI model provided the highest accuracy with 99% AUC for FFKC (75.00% sensitivity; 94.74% specificity).

Conclusions

Currently, AI parameters accurately diagnose AKC and EKC, but have a limited ability to diagnose FFKC. AI-assisted diagnostic technology that utilizes both SD-OCT and air-puff tonometry may overcome this limitation, leading to improved management of patients with KC.