{"id":776,"date":"2019-11-11T06:26:51","date_gmt":"2019-11-11T06:26:51","guid":{"rendered":"http:\/\/guires.uk\/newsroom\/?p=776"},"modified":"2019-11-27T12:10:14","modified_gmt":"2019-11-27T12:10:14","slug":"how-artificial-intelligence-enhances-the-detection-of-diabetic-retinopathy","status":"publish","type":"post","link":"https:\/\/guires.uk\/newsroom\/blog\/how-artificial-intelligence-enhances-the-detection-of-diabetic-retinopathy\/","title":{"rendered":"How Artificial Intelligence enhances the detection of Diabetic Retinopathy"},"content":{"rendered":"\n

Diabetic retinopathy (DR) is the leading causes of blindness and visual impairment worldwide. One in four Europeans over the age of 60 is affected by AMD, according to the report by EURETINA (European Society of Retina Specialist<\/a>). <\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

Yet, studies\nreported that 90% of the cases could be prevented through early detection and\ntreatment. Given the rising incidence of the cases, it’s challenging to\nmanually analyse the images and training new personnel is also a lengthy\nprocedure and requires acquisition of expertise. Deep learning (LeCun et al., 2015) (DL) a family of machine learning (ML)\ntechniques, has garnered attention recently because of its ability to learn the\nmost predictive features from images, given a large dataset of labelled\nexamples, without specifying rules or features explicitly. DL has been\nextensively used over the last several years for many automatic classification\ntasks, specifically, for the case of image classification. In the past, many DL\nclassifiers for DR have been published for diagnosis or prognosis, both of\nwhich involve predicting a label based on input data (AG, 2019). <\/p>\n\n\n\n

Several techniques\nhave been used in DR, including CNN, autoencoders (AEs), recurrent neural\nnetworks (RNNs), deep belief networks (DBNs). Despite several techniques, CNN\nfound to be most suitable for imaging data as it learns to perform its task\nthrough repetition and self-correction (Browne\n& Ghidary, 2003; Schmidhuber, 2015; LeCun et al., 2015). Recently studies have applied CNN in DL\nin the detection of referable diabetic retinopathy (Abr\u00e0moff et\nal., 2018; Ting et al., 2017a; Gulshan et al., 2016; Abr\u00e0moff et al., 2016;\nGargeya & Leng, 2017), glaucoma suspect (Ting et\nal., 2017b; Li et al., 2018), age-related macular degeneration (Ting et\nal., 2017b; Burlina et al., 2017; Grassmann et al., 2018) and retinopathy of prematurity (Brown et\nal., 2018). Gulshan et\nal. (Gulshan et\nal., 2016) developed CNN based DL system using 128175 macula-centred graded by a panel of ophthalmologists\nand 10000 images retrieved from two publicly available databases (EyePACS-1 and\nMessidor-2).  Authors in this study reported\n97.5% sensitivity and 93.4% specificity in the EYEPACS-1 while 96.1%\nsensitivity and 93.9% specificity for Messidor-2. <\/p>\n\n\n\n

Gargeya\nand Leng  (2017) developed diagnostic technology using deep\nconvolutional neural networks (based on the principles of deep residual\nlearning) to automate DR screening by processing color fundus images (75,137\nimages from the EyePACS public data), and classify as healthy or having DR. The\nmodel was tested using the public MESSIDOR 2 and E-Optha database for external\nvalidation. The developed model achieved 94% and 98% sensitivity and\nspecificity with 97% accuracy. Kermany et al.\n(2018) utilized transfer learning to classify\nage-related macular degeneration and diabetic macular oedema. The author had\nreported an accuracy of 96.6%, with a sensitivity of 97.8%, a specificity of\n97.4%. (de la Torre\net al., 2019) used receptive field DL classifier for detection\nof the most severe case of DR. The model was trained with 75,550 images. The\nmodel surpassed human expert capabilities, reaching the performance of 90% of\nsensitivity and specificity in test sets of about 10,000 images of patients. <\/p>\n\n\n\n

Earlier to train a\nDL algorithm, there is a requirement for human labeling of a reference training\nset. However, the study by Medeiros et al. (2019) used spectral-domain (SD) OCT data to\ntrain a DL algorithm to quantify glaucomatous structural damage on optic disc\nphotographs. The study used residual deep CNN to train and assess optic disc\nphotographs and predict SD-OCT average RNFL thickness. The author had reported\n94% specificity and sensitivity in discriminating glaucomatous from healthy\nsamples with the DL predicted and actual SD-OCT.  On the other hand, (Krause et\nal., 2018) trained CNN algorithm (Ensemble, hyper-parameter\nusing a Gaussian process bandit algorithm) using individual graders including\nUS board-certified ophthalmologists and retinal specialists. The kappa ranged\nfrom 82% to 91% accuracy in comparison to individual retinal specialists. <\/p>\n\n\n\n

But it is possible to prevent or delay if a high-risk group are identified at the right time and treated with the right approach. Machine learning algorithms do wonders where it helps to identify high risk population. <\/p>\n\n\n\n

<\/p>\n\n\n\n

Contact us for more information about how we can help you to develop CER report. <\/p>\n","protected":false},"excerpt":{"rendered":"

Diabetic retinopathy (DR) is the leading causes of blindness and visual impairment worldwide. One in four Europeans over the age of 60 is affected by AMD, according to the report by EURETINA (European Society of Retina Specialist). Yet, studies reported that 90% of the cases could be prevented through early detection and treatment. Given the […]<\/p>\n","protected":false},"author":1,"featured_media":925,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[53],"tags":[],"class_list":["post-776","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-blog"],"_links":{"self":[{"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/posts\/776","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/comments?post=776"}],"version-history":[{"count":4,"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/posts\/776\/revisions"}],"predecessor-version":[{"id":926,"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/posts\/776\/revisions\/926"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/media\/925"}],"wp:attachment":[{"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/media?parent=776"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/categories?post=776"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/guires.uk\/newsroom\/wp-json\/wp\/v2\/tags?post=776"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}