DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only

<p>Diabetes is one of the leading fatal diseases globally, putting a huge burden on the global healthcare system. Early diagnosis of diabetes is hence, of utmost importance and could save many lives. However, current techniques to determine whether a person has diabetes or has the risk of deve...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Mohammad Tariqul Islam (7854059) (author)
مؤلفون آخرون: Hamada R. H. Al-Absi (16726299) (author), Essam A. Ruagh (16896480) (author), Tanvir Alam (638619) (author)
منشور في: 2021
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513560456986624
author Mohammad Tariqul Islam (7854059)
author2 Hamada R. H. Al-Absi (16726299)
Essam A. Ruagh (16896480)
Tanvir Alam (638619)
author2_role author
author
author
author_facet Mohammad Tariqul Islam (7854059)
Hamada R. H. Al-Absi (16726299)
Essam A. Ruagh (16896480)
Tanvir Alam (638619)
author_role author
dc.creator.none.fl_str_mv Mohammad Tariqul Islam (7854059)
Hamada R. H. Al-Absi (16726299)
Essam A. Ruagh (16896480)
Tanvir Alam (638619)
dc.date.none.fl_str_mv 2021-01-18T00:00:00Z
dc.identifier.none.fl_str_mv 10.1109/access.2021.3052477
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/DiaNet_A_Deep_Learning_Based_Architecture_to_Diagnose_Diabetes_Using_Retinal_Images_Only/24049356
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biomedical and clinical sciences
Clinical sciences
Engineering
Biomedical engineering
Information and computing sciences
Machine learning
Diabetes
Retina
Medical diagnostic imaging
Retinopathy
Task analysis
Feature extraction
Convolutional neural network
Deep learning
Machine learning
Qatar
Qatar Biobank (QBB)
Magrabi Eye, Dental and Ear Centre
dc.title.none.fl_str_mv DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p>Diabetes is one of the leading fatal diseases globally, putting a huge burden on the global healthcare system. Early diagnosis of diabetes is hence, of utmost importance and could save many lives. However, current techniques to determine whether a person has diabetes or has the risk of developing diabetes are primarily reliant upon clinical biomarkers. In this article, we propose a novel deep learning architecture to predict if a person has diabetes or not from a photograph of his/her retina. Using a relatively small-sized dataset, we develop a multi-stage convolutional neural network (CNN)-based model DiaNet that can reach an accuracy level of over 84% on this task, and in doing so, successfully identifies the regions on the retina images that contribute to its decision-making process, as corroborated by the medical experts in the field. This is the first study that highlights the distinguishing capability of the retinal images for diabetes patients in the Qatari population to the best of our knowledge. Comparing the performance of DiaNet against the existing clinical data-based machine learning models, we conclude that the retinal images contain sufficient information to distinguish the Qatari diabetes cohort from the control group. In addition, our study reveals that retinal images may contain prognosis markers for diabetes and other comorbidities like hypertension and ischemic heart disease. The results led us to believe that the inclusion of retinal images into the clinical setup for the diagnosis of diabetes is warranted in the near future.</p><h2>Other Information</h2><p>Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/legalcode" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2021.3052477" target="_blank">https://dx.doi.org/10.1109/access.2021.3052477</a></p>
eu_rights_str_mv openAccess
id Manara2_9db6e81b6e75a1ab8e8547c1bea37332
identifier_str_mv 10.1109/access.2021.3052477
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/24049356
publishDate 2021
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images OnlyMohammad Tariqul Islam (7854059)Hamada R. H. Al-Absi (16726299)Essam A. Ruagh (16896480)Tanvir Alam (638619)Biomedical and clinical sciencesClinical sciencesEngineeringBiomedical engineeringInformation and computing sciencesMachine learningDiabetesRetinaMedical diagnostic imagingRetinopathyTask analysisFeature extractionConvolutional neural networkDeep learningMachine learningQatarQatar Biobank (QBB)Magrabi Eye, Dental and Ear Centre<p>Diabetes is one of the leading fatal diseases globally, putting a huge burden on the global healthcare system. Early diagnosis of diabetes is hence, of utmost importance and could save many lives. However, current techniques to determine whether a person has diabetes or has the risk of developing diabetes are primarily reliant upon clinical biomarkers. In this article, we propose a novel deep learning architecture to predict if a person has diabetes or not from a photograph of his/her retina. Using a relatively small-sized dataset, we develop a multi-stage convolutional neural network (CNN)-based model DiaNet that can reach an accuracy level of over 84% on this task, and in doing so, successfully identifies the regions on the retina images that contribute to its decision-making process, as corroborated by the medical experts in the field. This is the first study that highlights the distinguishing capability of the retinal images for diabetes patients in the Qatari population to the best of our knowledge. Comparing the performance of DiaNet against the existing clinical data-based machine learning models, we conclude that the retinal images contain sufficient information to distinguish the Qatari diabetes cohort from the control group. In addition, our study reveals that retinal images may contain prognosis markers for diabetes and other comorbidities like hypertension and ischemic heart disease. The results led us to believe that the inclusion of retinal images into the clinical setup for the diagnosis of diabetes is warranted in the near future.</p><h2>Other Information</h2><p>Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/legalcode" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2021.3052477" target="_blank">https://dx.doi.org/10.1109/access.2021.3052477</a></p>2021-01-18T00:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/access.2021.3052477https://figshare.com/articles/journal_contribution/DiaNet_A_Deep_Learning_Based_Architecture_to_Diagnose_Diabetes_Using_Retinal_Images_Only/24049356CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/240493562021-01-18T00:00:00Z
spellingShingle DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
Mohammad Tariqul Islam (7854059)
Biomedical and clinical sciences
Clinical sciences
Engineering
Biomedical engineering
Information and computing sciences
Machine learning
Diabetes
Retina
Medical diagnostic imaging
Retinopathy
Task analysis
Feature extraction
Convolutional neural network
Deep learning
Machine learning
Qatar
Qatar Biobank (QBB)
Magrabi Eye, Dental and Ear Centre
status_str publishedVersion
title DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
title_full DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
title_fullStr DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
title_full_unstemmed DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
title_short DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
title_sort DiaNet: A Deep Learning Based Architecture to Diagnose Diabetes Using Retinal Images Only
topic Biomedical and clinical sciences
Clinical sciences
Engineering
Biomedical engineering
Information and computing sciences
Machine learning
Diabetes
Retina
Medical diagnostic imaging
Retinopathy
Task analysis
Feature extraction
Convolutional neural network
Deep learning
Machine learning
Qatar
Qatar Biobank (QBB)
Magrabi Eye, Dental and Ear Centre