ePoster

Die hochgeladenen ePoster sind nur für eingeloggte Teilnehmende ersichtlich.

Bitte loggen Sie sich ein, um zu den ePoster zu gelangen.

Detaillierte Beitrags-Information

Beitragstitel	Accuracy of autonomous artificial intelligence-based diabetic retinopathy screening in real life clinical practice
Beitragscode	P61
Autor:innen	Eleonora Riotto hôpital Jules Gonin, hôpital ophtalmique Präsentierende:r Stefan Gasser Jules Gonin Eye Hospital Jelena Potic Hôpital Ophtlamique Jules Gonin, Fondation Asile des Aveugles, Université de Lausanne Mohamed Sherif HOPITAL JULES GONIN - Fondation Asile des Aveugles Theodor Stappler Hopital Jules Gonin, Fondation asile des aveugles Reinier Schlingemann Hôpital Ophtalmique Jules Gonin, Université de Lausanne Thomas J. Wolfensberger Hôpital ophtalmique Jules-Gonin Lazaros Konstantinidis Jules-Gonin Eye Hospital, University of Lausanne, Switzerland
Präsentationsform	ePoster
Themengebiete	Retina Vitreous
Abstract-Text	Introduction In diabetic retinopathy early detection and intervention are crucial in preventing vision loss and improving patient outcomes. In the era of artificial intelligence (AI) and machine learning new promising diagnostic tools have emerged. The IDX-DR machine (Digital Diagnostics, Coralville, IA, USA) represents a diagnostic tool that combines advanced imaging techniques, AI algorithms, and deep learning methodologies to identify and classify diabetic retinopathy. Methods All patients that participated to our AI-based DR screening were considered for this study. For this study all retinal images were additionally reviewed retrospectively by 2 experienced retinal specialists. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated for the IDX-DR machine compared to the graders’ responses. Results We included a total of 2282 images from 1141 patients that were screened between January 2021 and April 2023 at the Jules Gonin Eye Hospital in Lausanne, Switzerland. Sensitivity was calculated to be 100% for ‘no DR’, ‘mild DR’ and ‘moderate DR’. Specificity for no DR’, ‘mild DR’, ‘moderate DR’ and ‘severe DR’ was calculated to be respectively 78.4%, 81.2%, 93.4% and 97.6%. PPV was calculated to be respectively 36.7%, 24.6%, 1.4% and 0%. NPV was calculated to be 100% for each category. Accuracy was calculated to be higher than 80% for ‘no DR’, ‘mild DR’ and ‘moderate DR’. Discussion In this study based in Jules Gonin Eye Hospital in Lausanne, we compared the autonomous diagnostic AI system of IDX-DR machine to detect diabetic retinopathy to a human grading established by two experienced retinal specialists. Our results showed that ID-x DR machine constantly overestimate the DR stages thus permitting the clinicians to fully trust negative results delivered by the screening software. Nevertheless, all fundus of images classified as a ‘mild DR’ or greater should always be controlled by a specialist in order to assert if the predicted stage is truly present.