Assessment of patient specific information in the wild on fundus photography and optical coherence tomography

Abstract In this paper we analyse the performance of machine learning methods in predicting patient information such as age or sex solely from retinal imaging modalities in a heterogeneous clinical population. Our dataset consists of N = 135,667 fundus images and N = 85,536 volumetric OCT scans. Dee...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Marion R. Munk, Thomas Kurmann, Pablo Márquez-Neila, Martin S. Zinkernagel, Sebastian Wolf, Raphael Sznitman
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/6b98e74c81494d849646e7f54ff22d73
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:6b98e74c81494d849646e7f54ff22d73
record_format	dspace
spelling	oai:doaj.org-article:6b98e74c81494d849646e7f54ff22d732021-12-02T13:39:47ZAssessment of patient specific information in the wild on fundus photography and optical coherence tomography10.1038/s41598-021-86577-52045-2322https://doaj.org/article/6b98e74c81494d849646e7f54ff22d732021-04-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-86577-5https://doaj.org/toc/2045-2322Abstract In this paper we analyse the performance of machine learning methods in predicting patient information such as age or sex solely from retinal imaging modalities in a heterogeneous clinical population. Our dataset consists of N = 135,667 fundus images and N = 85,536 volumetric OCT scans. Deep learning models were trained to predict the patient’s age and sex from fundus images, OCT cross sections and OCT volumes. For sex prediction, a ROC AUC of 0.80 was achieved for fundus images, 0.84 for OCT cross sections and 0.90 for OCT volumes. Age prediction mean absolute errors of 6.328 years for fundus, 5.625 years for OCT cross sections and 4.541 for OCT volumes were observed. We assess the performance of OCT scans containing different biomarkers and note a peak performance of AUC = 0.88 for OCT cross sections and 0.95 for volumes when there is no pathology on scans. Performance drops in case of drusen, fibrovascular pigment epitheliuum detachment and geographic atrophy present. We conclude that deep learning based methods are capable of classifying the patient’s sex and age from color fundus photography and OCT for a broad spectrum of patients irrespective of underlying disease or image quality. Non-random sex prediction using fundus images seems only possible if the eye fovea and optic disc are visible.Marion R. MunkThomas KurmannPablo Márquez-NeilaMartin S. ZinkernagelSebastian WolfRaphael SznitmanNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-10 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Marion R. Munk Thomas Kurmann Pablo Márquez-Neila Martin S. Zinkernagel Sebastian Wolf Raphael Sznitman Assessment of patient specific information in the wild on fundus photography and optical coherence tomography
description	Abstract In this paper we analyse the performance of machine learning methods in predicting patient information such as age or sex solely from retinal imaging modalities in a heterogeneous clinical population. Our dataset consists of N = 135,667 fundus images and N = 85,536 volumetric OCT scans. Deep learning models were trained to predict the patient’s age and sex from fundus images, OCT cross sections and OCT volumes. For sex prediction, a ROC AUC of 0.80 was achieved for fundus images, 0.84 for OCT cross sections and 0.90 for OCT volumes. Age prediction mean absolute errors of 6.328 years for fundus, 5.625 years for OCT cross sections and 4.541 for OCT volumes were observed. We assess the performance of OCT scans containing different biomarkers and note a peak performance of AUC = 0.88 for OCT cross sections and 0.95 for volumes when there is no pathology on scans. Performance drops in case of drusen, fibrovascular pigment epitheliuum detachment and geographic atrophy present. We conclude that deep learning based methods are capable of classifying the patient’s sex and age from color fundus photography and OCT for a broad spectrum of patients irrespective of underlying disease or image quality. Non-random sex prediction using fundus images seems only possible if the eye fovea and optic disc are visible.
format	article
author	Marion R. Munk Thomas Kurmann Pablo Márquez-Neila Martin S. Zinkernagel Sebastian Wolf Raphael Sznitman
author_facet	Marion R. Munk Thomas Kurmann Pablo Márquez-Neila Martin S. Zinkernagel Sebastian Wolf Raphael Sznitman
author_sort	Marion R. Munk
title	Assessment of patient specific information in the wild on fundus photography and optical coherence tomography
title_short	Assessment of patient specific information in the wild on fundus photography and optical coherence tomography
title_full	Assessment of patient specific information in the wild on fundus photography and optical coherence tomography
title_fullStr	Assessment of patient specific information in the wild on fundus photography and optical coherence tomography
title_full_unstemmed	Assessment of patient specific information in the wild on fundus photography and optical coherence tomography
title_sort	assessment of patient specific information in the wild on fundus photography and optical coherence tomography
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/6b98e74c81494d849646e7f54ff22d73
work_keys_str_mv	AT marionrmunk assessmentofpatientspecificinformationinthewildonfundusphotographyandopticalcoherencetomography AT thomaskurmann assessmentofpatientspecificinformationinthewildonfundusphotographyandopticalcoherencetomography AT pablomarquezneila assessmentofpatientspecificinformationinthewildonfundusphotographyandopticalcoherencetomography AT martinszinkernagel assessmentofpatientspecificinformationinthewildonfundusphotographyandopticalcoherencetomography AT sebastianwolf assessmentofpatientspecificinformationinthewildonfundusphotographyandopticalcoherencetomography AT raphaelsznitman assessmentofpatientspecificinformationinthewildonfundusphotographyandopticalcoherencetomography
_version_	1718392620716130304

Assessment of patient specific information in the wild on fundus photography and optical coherence tomography

Ejemplares similares