Identifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning

The aim of this study was to develop and validate a deep learning-based system to detect peripheral neuropathy (DN) from retinal colour images in people with diabetes. Retinal images from 1561 people with diabetes were used to predictDN diagnosed on vibration perception threshold. A total of 189 had...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Diego R. Cervera, Luke Smith, Luis Diaz-Santana, Meenakshi Kumar, Rajiv Raman, Sobha Sivaprasad
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
Acceso en línea:https://doaj.org/article/ef0f54e0eb31403fae121aa4516d6668
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:ef0f54e0eb31403fae121aa4516d6668
record_format dspace
spelling oai:doaj.org-article:ef0f54e0eb31403fae121aa4516d66682021-11-25T17:20:06ZIdentifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning10.3390/diagnostics111119432075-4418https://doaj.org/article/ef0f54e0eb31403fae121aa4516d66682021-10-01T00:00:00Zhttps://www.mdpi.com/2075-4418/11/11/1943https://doaj.org/toc/2075-4418The aim of this study was to develop and validate a deep learning-based system to detect peripheral neuropathy (DN) from retinal colour images in people with diabetes. Retinal images from 1561 people with diabetes were used to predictDN diagnosed on vibration perception threshold. A total of 189 had diabetic retinopathy (DR), 276 had DN, and 43 had both DR and DN. 90% of the images were used for training and validation and 10% for testing. Deep neural networks, including Squeezenet, Inception, and Densenet were utilized, and the architectures were tested with and without pre-trained weights. Random transform of images was used during training. The algorithm was trained and tested using three sets of data: all retinal images, images without DR and images with DR. Area under the ROC curve (AUC) was used to evaluate performance. The AUC to predict DN on the whole cohort was 0.8013 (±0.0257) on the validation set and 0.7097 (±0.0031) on the test set. The AUC increased to 0.8673 (±0.0088) in the presence of DR. The retinal images can be used to identify individuals with DN and provides an opportunity to educate patients about their DN status when they attend DR screening.Diego R. CerveraLuke SmithLuis Diaz-SantanaMeenakshi KumarRajiv RamanSobha SivaprasadMDPI AGarticlediabetesdeep learningdiabetic neuropathydiabetic retinopathyMedicine (General)R5-920ENDiagnostics, Vol 11, Iss 1943, p 1943 (2021)
institution DOAJ
collection DOAJ
language EN
topic diabetes
deep learning
diabetic neuropathy
diabetic retinopathy
Medicine (General)
R5-920
spellingShingle diabetes
deep learning
diabetic neuropathy
diabetic retinopathy
Medicine (General)
R5-920
Diego R. Cervera
Luke Smith
Luis Diaz-Santana
Meenakshi Kumar
Rajiv Raman
Sobha Sivaprasad
Identifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning
description The aim of this study was to develop and validate a deep learning-based system to detect peripheral neuropathy (DN) from retinal colour images in people with diabetes. Retinal images from 1561 people with diabetes were used to predictDN diagnosed on vibration perception threshold. A total of 189 had diabetic retinopathy (DR), 276 had DN, and 43 had both DR and DN. 90% of the images were used for training and validation and 10% for testing. Deep neural networks, including Squeezenet, Inception, and Densenet were utilized, and the architectures were tested with and without pre-trained weights. Random transform of images was used during training. The algorithm was trained and tested using three sets of data: all retinal images, images without DR and images with DR. Area under the ROC curve (AUC) was used to evaluate performance. The AUC to predict DN on the whole cohort was 0.8013 (±0.0257) on the validation set and 0.7097 (±0.0031) on the test set. The AUC increased to 0.8673 (±0.0088) in the presence of DR. The retinal images can be used to identify individuals with DN and provides an opportunity to educate patients about their DN status when they attend DR screening.
format article
author Diego R. Cervera
Luke Smith
Luis Diaz-Santana
Meenakshi Kumar
Rajiv Raman
Sobha Sivaprasad
author_facet Diego R. Cervera
Luke Smith
Luis Diaz-Santana
Meenakshi Kumar
Rajiv Raman
Sobha Sivaprasad
author_sort Diego R. Cervera
title Identifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning
title_short Identifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning
title_full Identifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning
title_fullStr Identifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning
title_full_unstemmed Identifying Peripheral Neuropathy in Colour Fundus Photographs Based on Deep Learning
title_sort identifying peripheral neuropathy in colour fundus photographs based on deep learning
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/ef0f54e0eb31403fae121aa4516d6668
work_keys_str_mv AT diegorcervera identifyingperipheralneuropathyincolourfundusphotographsbasedondeeplearning
AT lukesmith identifyingperipheralneuropathyincolourfundusphotographsbasedondeeplearning
AT luisdiazsantana identifyingperipheralneuropathyincolourfundusphotographsbasedondeeplearning
AT meenakshikumar identifyingperipheralneuropathyincolourfundusphotographsbasedondeeplearning
AT rajivraman identifyingperipheralneuropathyincolourfundusphotographsbasedondeeplearning
AT sobhasivaprasad identifyingperipheralneuropathyincolourfundusphotographsbasedondeeplearning
_version_ 1718412504956141568