Cargando…

Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK

PURPOSE: To evaluate a novel deep learning algorithm to distinguish between eyes that may or may not have a graft detachment based on pre–Descemet membrane endothelial keratoplasty (DMEK) anterior segment optical coherence tomography (AS-OCT) images. METHODS: Retrospective cohort study. A multiple-i...

Descripción completa

Detalles Bibliográficos
Autores principales: Patefield, Alastair, Meng, Yanda, Airaldi, Matteo, Coco, Giulia, Vaccaro, Sabrina, Parekh, Mohit, Semeraro, Francesco, Gadhvi, Kunal A., Kaye, Stephen B., Zheng, Yalin, Romano, Vito
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Association for Research in Vision and Ophthalmology 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10187791/
https://www.ncbi.nlm.nih.gov/pubmed/37184500
http://dx.doi.org/10.1167/tvst.12.5.14
_version_ 1785042797465698304
author Patefield, Alastair
Meng, Yanda
Airaldi, Matteo
Coco, Giulia
Vaccaro, Sabrina
Parekh, Mohit
Semeraro, Francesco
Gadhvi, Kunal A.
Kaye, Stephen B.
Zheng, Yalin
Romano, Vito
author_facet Patefield, Alastair
Meng, Yanda
Airaldi, Matteo
Coco, Giulia
Vaccaro, Sabrina
Parekh, Mohit
Semeraro, Francesco
Gadhvi, Kunal A.
Kaye, Stephen B.
Zheng, Yalin
Romano, Vito
author_sort Patefield, Alastair
collection PubMed
description PURPOSE: To evaluate a novel deep learning algorithm to distinguish between eyes that may or may not have a graft detachment based on pre–Descemet membrane endothelial keratoplasty (DMEK) anterior segment optical coherence tomography (AS-OCT) images. METHODS: Retrospective cohort study. A multiple-instance learning artificial intelligence (MIL-AI) model using a ResNet-101 backbone was designed. AS-OCT images were split into training and testing sets. The MIL-AI model was trained and validated on the training set. Model performance and heatmaps were calculated from the testing set. Classification performance metrics included F1 score (harmonic mean of recall and precision), specificity, sensitivity, and area under curve (AUC). Finally, MIL-AI performance was compared to manual classification by an experienced ophthalmologist. RESULTS: In total, 9466 images of 74 eyes (128 images per eye) were included in the study. Images from 50 eyes were used to train and validate the MIL-AI system, while the remaining 24 eyes were used as the test set to determine its performance and generate heatmaps for visualization. The performance metrics on the test set (95% confidence interval) were as follows: F1 score, 0.77 (0.57–0.91); precision, 0.67 (0.44–0.88); specificity, 0.45 (0.15–0.75); sensitivity, 0.92 (0.73–1.00); and AUC, 0.63 (0.52–0.86). MIL-AI performance was more sensitive (92% vs. 31%) but less specific (45% vs. 64%) than the ophthalmologist's performance. CONCLUSIONS: The MIL-AI predicts with high sensitivity the eyes that may have post-DMEK graft detachment requiring rebubbling. Larger-scale clinical trials are warranted to validate the model. TRANSLATIONAL RELEVANCE: MIL-AI models represent an opportunity for implementation in routine DMEK suitability screening.
format Online
Article
Text
id pubmed-10187791
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher The Association for Research in Vision and Ophthalmology
record_format MEDLINE/PubMed
spelling pubmed-101877912023-05-17 Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK Patefield, Alastair Meng, Yanda Airaldi, Matteo Coco, Giulia Vaccaro, Sabrina Parekh, Mohit Semeraro, Francesco Gadhvi, Kunal A. Kaye, Stephen B. Zheng, Yalin Romano, Vito Transl Vis Sci Technol Artificial Intelligence PURPOSE: To evaluate a novel deep learning algorithm to distinguish between eyes that may or may not have a graft detachment based on pre–Descemet membrane endothelial keratoplasty (DMEK) anterior segment optical coherence tomography (AS-OCT) images. METHODS: Retrospective cohort study. A multiple-instance learning artificial intelligence (MIL-AI) model using a ResNet-101 backbone was designed. AS-OCT images were split into training and testing sets. The MIL-AI model was trained and validated on the training set. Model performance and heatmaps were calculated from the testing set. Classification performance metrics included F1 score (harmonic mean of recall and precision), specificity, sensitivity, and area under curve (AUC). Finally, MIL-AI performance was compared to manual classification by an experienced ophthalmologist. RESULTS: In total, 9466 images of 74 eyes (128 images per eye) were included in the study. Images from 50 eyes were used to train and validate the MIL-AI system, while the remaining 24 eyes were used as the test set to determine its performance and generate heatmaps for visualization. The performance metrics on the test set (95% confidence interval) were as follows: F1 score, 0.77 (0.57–0.91); precision, 0.67 (0.44–0.88); specificity, 0.45 (0.15–0.75); sensitivity, 0.92 (0.73–1.00); and AUC, 0.63 (0.52–0.86). MIL-AI performance was more sensitive (92% vs. 31%) but less specific (45% vs. 64%) than the ophthalmologist's performance. CONCLUSIONS: The MIL-AI predicts with high sensitivity the eyes that may have post-DMEK graft detachment requiring rebubbling. Larger-scale clinical trials are warranted to validate the model. TRANSLATIONAL RELEVANCE: MIL-AI models represent an opportunity for implementation in routine DMEK suitability screening. The Association for Research in Vision and Ophthalmology 2023-05-15 /pmc/articles/PMC10187791/ /pubmed/37184500 http://dx.doi.org/10.1167/tvst.12.5.14 Text en Copyright 2023 The Authors https://creativecommons.org/licenses/by/4.0/This work is licensed under a Creative Commons Attribution 4.0 International License.
spellingShingle Artificial Intelligence
Patefield, Alastair
Meng, Yanda
Airaldi, Matteo
Coco, Giulia
Vaccaro, Sabrina
Parekh, Mohit
Semeraro, Francesco
Gadhvi, Kunal A.
Kaye, Stephen B.
Zheng, Yalin
Romano, Vito
Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK
title Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK
title_full Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK
title_fullStr Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK
title_full_unstemmed Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK
title_short Deep Learning Using Preoperative AS-OCT Predicts Graft Detachment in DMEK
title_sort deep learning using preoperative as-oct predicts graft detachment in dmek
topic Artificial Intelligence
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10187791/
https://www.ncbi.nlm.nih.gov/pubmed/37184500
http://dx.doi.org/10.1167/tvst.12.5.14
work_keys_str_mv AT patefieldalastair deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT mengyanda deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT airaldimatteo deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT cocogiulia deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT vaccarosabrina deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT parekhmohit deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT semerarofrancesco deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT gadhvikunala deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT kayestephenb deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT zhengyalin deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek
AT romanovito deeplearningusingpreoperativeasoctpredictsgraftdetachmentindmek