Cargando…

A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography

PURPOSE: The purpose of this study was to develop and validate a deep learning (DL) framework for the detection and quantification of reticular pseudodrusen (RPD) and drusen on optical coherence tomography (OCT) scans. METHODS: A DL framework was developed consisting of a classification model and an...

Descripción completa

Detalles Bibliográficos
Autores principales: Schwartz, Roy, Khalid, Hagar, Liakopoulos, Sandra, Ouyang, Yanling, de Vente, Coen, González-Gonzalo, Cristina, Lee, Aaron Y., Guymer, Robyn, Chew, Emily Y., Egan, Catherine, Wu, Zhichao, Kumar, Himeesh, Farrington, Joseph, Müller, Philipp L., Sánchez, Clara I., Tufail, Adnan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Association for Research in Vision and Ophthalmology 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9728496/
https://www.ncbi.nlm.nih.gov/pubmed/36458946
http://dx.doi.org/10.1167/tvst.11.12.3
_version_ 1784845269311946752
author Schwartz, Roy
Khalid, Hagar
Liakopoulos, Sandra
Ouyang, Yanling
de Vente, Coen
González-Gonzalo, Cristina
Lee, Aaron Y.
Guymer, Robyn
Chew, Emily Y.
Egan, Catherine
Wu, Zhichao
Kumar, Himeesh
Farrington, Joseph
Müller, Philipp L.
Sánchez, Clara I.
Tufail, Adnan
author_facet Schwartz, Roy
Khalid, Hagar
Liakopoulos, Sandra
Ouyang, Yanling
de Vente, Coen
González-Gonzalo, Cristina
Lee, Aaron Y.
Guymer, Robyn
Chew, Emily Y.
Egan, Catherine
Wu, Zhichao
Kumar, Himeesh
Farrington, Joseph
Müller, Philipp L.
Sánchez, Clara I.
Tufail, Adnan
author_sort Schwartz, Roy
collection PubMed
description PURPOSE: The purpose of this study was to develop and validate a deep learning (DL) framework for the detection and quantification of reticular pseudodrusen (RPD) and drusen on optical coherence tomography (OCT) scans. METHODS: A DL framework was developed consisting of a classification model and an out-of-distribution (OOD) detection model for the identification of ungradable scans; a classification model to identify scans with drusen or RPD; and an image segmentation model to independently segment lesions as RPD or drusen. Data were obtained from 1284 participants in the UK Biobank (UKBB) with a self-reported diagnosis of age-related macular degeneration (AMD) and 250 UKBB controls. Drusen and RPD were manually delineated by five retina specialists. The main outcome measures were sensitivity, specificity, area under the receiver operating characteristic (ROC) curve (AUC), kappa, accuracy, intraclass correlation coefficient (ICC), and free-response receiver operating characteristic (FROC) curves. RESULTS: The classification models performed strongly at their respective tasks (0.95, 0.93, and 0.99 AUC, respectively, for the ungradable scans classifier, the OOD model, and the drusen and RPD classification models). The mean ICC for the drusen and RPD area versus graders was 0.74 and 0.61, respectively, compared with 0.69 and 0.68 for intergrader agreement. FROC curves showed that the model's sensitivity was close to human performance. CONCLUSIONS: The models achieved high classification and segmentation performance, similar to human performance. TRANSLATIONAL RELEVANCE: Application of this robust framework will further our understanding of RPD as a separate entity from drusen in both research and clinical settings.
format Online
Article
Text
id pubmed-9728496
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher The Association for Research in Vision and Ophthalmology
record_format MEDLINE/PubMed
spelling pubmed-97284962022-12-08 A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography Schwartz, Roy Khalid, Hagar Liakopoulos, Sandra Ouyang, Yanling de Vente, Coen González-Gonzalo, Cristina Lee, Aaron Y. Guymer, Robyn Chew, Emily Y. Egan, Catherine Wu, Zhichao Kumar, Himeesh Farrington, Joseph Müller, Philipp L. Sánchez, Clara I. Tufail, Adnan Transl Vis Sci Technol Artificial Intelligence PURPOSE: The purpose of this study was to develop and validate a deep learning (DL) framework for the detection and quantification of reticular pseudodrusen (RPD) and drusen on optical coherence tomography (OCT) scans. METHODS: A DL framework was developed consisting of a classification model and an out-of-distribution (OOD) detection model for the identification of ungradable scans; a classification model to identify scans with drusen or RPD; and an image segmentation model to independently segment lesions as RPD or drusen. Data were obtained from 1284 participants in the UK Biobank (UKBB) with a self-reported diagnosis of age-related macular degeneration (AMD) and 250 UKBB controls. Drusen and RPD were manually delineated by five retina specialists. The main outcome measures were sensitivity, specificity, area under the receiver operating characteristic (ROC) curve (AUC), kappa, accuracy, intraclass correlation coefficient (ICC), and free-response receiver operating characteristic (FROC) curves. RESULTS: The classification models performed strongly at their respective tasks (0.95, 0.93, and 0.99 AUC, respectively, for the ungradable scans classifier, the OOD model, and the drusen and RPD classification models). The mean ICC for the drusen and RPD area versus graders was 0.74 and 0.61, respectively, compared with 0.69 and 0.68 for intergrader agreement. FROC curves showed that the model's sensitivity was close to human performance. CONCLUSIONS: The models achieved high classification and segmentation performance, similar to human performance. TRANSLATIONAL RELEVANCE: Application of this robust framework will further our understanding of RPD as a separate entity from drusen in both research and clinical settings. The Association for Research in Vision and Ophthalmology 2022-12-02 /pmc/articles/PMC9728496/ /pubmed/36458946 http://dx.doi.org/10.1167/tvst.11.12.3 Text en Copyright 2022 The Authors https://creativecommons.org/licenses/by-nc-nd/4.0/This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
spellingShingle Artificial Intelligence
Schwartz, Roy
Khalid, Hagar
Liakopoulos, Sandra
Ouyang, Yanling
de Vente, Coen
González-Gonzalo, Cristina
Lee, Aaron Y.
Guymer, Robyn
Chew, Emily Y.
Egan, Catherine
Wu, Zhichao
Kumar, Himeesh
Farrington, Joseph
Müller, Philipp L.
Sánchez, Clara I.
Tufail, Adnan
A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography
title A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography
title_full A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography
title_fullStr A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography
title_full_unstemmed A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography
title_short A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography
title_sort deep learning framework for the detection and quantification of reticular pseudodrusen and drusen on optical coherence tomography
topic Artificial Intelligence
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9728496/
https://www.ncbi.nlm.nih.gov/pubmed/36458946
http://dx.doi.org/10.1167/tvst.11.12.3
work_keys_str_mv AT schwartzroy adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT khalidhagar adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT liakopoulossandra adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT ouyangyanling adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT deventecoen adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT gonzalezgonzalocristina adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT leeaarony adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT guymerrobyn adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT chewemilyy adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT egancatherine adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT wuzhichao adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT kumarhimeesh adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT farringtonjoseph adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT mullerphilippl adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT sanchezclarai adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT tufailadnan adeeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT schwartzroy deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT khalidhagar deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT liakopoulossandra deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT ouyangyanling deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT deventecoen deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT gonzalezgonzalocristina deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT leeaarony deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT guymerrobyn deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT chewemilyy deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT egancatherine deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT wuzhichao deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT kumarhimeesh deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT farringtonjoseph deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT mullerphilippl deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT sanchezclarai deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography
AT tufailadnan deeplearningframeworkforthedetectionandquantificationofreticularpseudodrusenanddrusenonopticalcoherencetomography