Type presentatie | Oral presentation |
Titel | Training the “Cone Location and Magnitude Index Expanded” on Keratoconus Suspect Eyes using Machine Learning |
Doel | To use artificial intelligence in retraining the Cone Location and Magnitude Index Expanded to detect keratoconus suspect corneas. |
Methodes | 349 eyes of 349 patients from the American University of Beirut Medical Center were included in the study, with 133 eyes with normal control corneas (NL), 147 eyes with definite keratoconus (KC), and 75 eyes with keratoconus suspect corneas (KCS). KCS corneas were normal fellow eyes of patients with highly asymmetric keratoconus. Different parameters from anterior, posterior corneal map and pachymetry were used. |
Resultaten | Using logistic regression, the machine learning model trained on three categories of eyes achieved a sensitivity of 67%, specificity of 87%, positive predictive value (PPV) of 58%, and F1 score of 0.62 in detecting KCS. A new artificial intelligence algorithm (CatBoost) scored 71% of sensitivity, 91% of specificity, a PPV of 68% and F1 score of 0.70, while using the same parameters incorporated in the original index. |
Conclusie | Artificial intelligence improves the performance of the Cone Location and Magnitude Index Expanded in detecting early suspicion of keratoconus. |
Belangenverstrengeling | Nee |
Details verstrengelde belangen | None |
Naam | WEHBI |
Initialen | Z |
Instituut | American University of Beirut |
Stad | Beirut, Lebanon |
Naam | Hammoud |
Initialen | B |
Instituut | American University of Beirut |
Stad | Beirut, Lebanon |
Naam | Assaf |
Initialen | J.F |
Instituut | Casey Eye Institute |
Stad | Portland, Oregon |
Naam | Awwad |
Initialen | S.T |
Instituut | American University of Beirut |
Stad | Beirut, Lebanon |