TY - JOUR
T1 - Classification of diabetes-related retinal diseases using a deep learning approach in optical coherence tomography
AU - Perdomo, Oscar
AU - Rios, Hernán
AU - Rodríguez, Francisco J.
AU - Otálora, Sebastián
AU - Meriaudeau, Fabrice
AU - Müller, Henning
AU - González, Fabio A.
N1 - Funding Information:
This work was funded by the project Detección temprana de daño ocular en diabéticos usando un sistema de inteligencia artificial en imágenes de fondo de ojo number 1101-807-63563 CT of Colciencias by Convocatoria Colciencias 837 de 2018. Oscar Perdomo thanks COLCIENCIAS for funding this research with a doctoral grant. Sebastian Otálora thanks Colciencias for funding partially this research with a doctoral grant through the call 756 for PhD. programs. We appreciate the efforts devoted by: Prof. Sina Farsiu and Prof. Cynthia A. Toth from Duke University; Carol Cheung and Tien Y Wong from the Chinese University of Hong Kong (CUHK) and Singapore Eye Research Institute (SERI) to collect SD-OCT volumes. This work was partially supported by Nvidia with a TitanX GPU .
Publisher Copyright:
© 2019 Elsevier B.V.
Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.
PY - 2019/9
Y1 - 2019/9
N2 - Background and objectives: Spectral Domain Optical Coherence Tomography (SD-OCT) is a volumetric imaging technique that allows measuring patterns between layers such as small amounts of fluid. Since 2012, automatic medical image analysis performance has steadily increased through the use of deep learning models that automatically learn relevant features for specific tasks, instead of designing visual features manually. Nevertheless, providing insights and interpretation of the predictions made by the model is still a challenge. This paper describes a deep learning model able to detect medically interpretable information in relevant images from a volume to classify diabetes-related retinal diseases. Methods: This article presents a new deep learning model, OCT-NET, which is a customized convolutional neural network for processing scans extracted from optical coherence tomography volumes. OCT-NET is applied to the classification of three conditions seen in SD-OCT volumes. Additionally, the proposed model includes a feedback stage that highlights the areas of the scans to support the interpretation of the results. This information is potentially useful for a medical specialist while assessing the prediction produced by the model. Results: The proposed model was tested on the public SERI-CUHK and A2A SD-OCT data sets containing healthy, diabetic retinopathy, diabetic macular edema and age-related macular degeneration. The experimental evaluation shows that the proposed method outperforms conventional convolutional deep learning models from the state of the art reported on the SERI+CUHK and A2A SD-OCT data sets with a precision of 93% and an area under the ROC curve (AUC) of 0.99 respectively. Conclusions: The proposed method is able to classify the three studied retinal diseases with high accuracy. One advantage of the method is its ability to produce interpretable clinical information in the form of highlighting the regions of the image that most contribute to the classifier decision.
AB - Background and objectives: Spectral Domain Optical Coherence Tomography (SD-OCT) is a volumetric imaging technique that allows measuring patterns between layers such as small amounts of fluid. Since 2012, automatic medical image analysis performance has steadily increased through the use of deep learning models that automatically learn relevant features for specific tasks, instead of designing visual features manually. Nevertheless, providing insights and interpretation of the predictions made by the model is still a challenge. This paper describes a deep learning model able to detect medically interpretable information in relevant images from a volume to classify diabetes-related retinal diseases. Methods: This article presents a new deep learning model, OCT-NET, which is a customized convolutional neural network for processing scans extracted from optical coherence tomography volumes. OCT-NET is applied to the classification of three conditions seen in SD-OCT volumes. Additionally, the proposed model includes a feedback stage that highlights the areas of the scans to support the interpretation of the results. This information is potentially useful for a medical specialist while assessing the prediction produced by the model. Results: The proposed model was tested on the public SERI-CUHK and A2A SD-OCT data sets containing healthy, diabetic retinopathy, diabetic macular edema and age-related macular degeneration. The experimental evaluation shows that the proposed method outperforms conventional convolutional deep learning models from the state of the art reported on the SERI+CUHK and A2A SD-OCT data sets with a precision of 93% and an area under the ROC curve (AUC) of 0.99 respectively. Conclusions: The proposed method is able to classify the three studied retinal diseases with high accuracy. One advantage of the method is its ability to produce interpretable clinical information in the form of highlighting the regions of the image that most contribute to the classifier decision.
UR - http://www.scopus.com/inward/record.url?scp=85068138206&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85068138206&partnerID=8YFLogxK
U2 - 10.1016/j.cmpb.2019.06.016
DO - 10.1016/j.cmpb.2019.06.016
M3 - Article
C2 - 31416547
AN - SCOPUS:85068138206
SN - 0169-2607
VL - 178
SP - 18
EP - 189
JO - Computer Methods and Programs in Biomedicine
JF - Computer Methods and Programs in Biomedicine
ER -