Deep learning in ophthalmology: a review

Grewal, Parampal S.; Oloumi, Faraz; Rubin, Uriel; Tennant, Matthew

doi:10.1016/j.jcjo.2018.04.019

Cited by 112 publications

(54 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Incorporating AI to accurately grade the retinal images for DR would offer many benefits to DR screening programs; reducing their reliance on trained clinicians to read photographs, enabling point of contact diagnosis whilst reducing the need for complex IT support systems. Research into AI design and its development for DR screening has progressed significantly in recent years, and this field has enjoyed a good deal of attention of late [60][61][62]. However, for all the excitement very little of this work has progressed to a clinically useful tool which provides a real-world AI-solution for DR screening programs and this is due largely to the challenges of the research-driven AI to generalize to a real-world setup.…”

Section: Plos Onementioning

confidence: 99%

Towards implementation of AI in New Zealand national diabetic screening program: Cloud-based, robust, and bespoke

et al. 2020

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) have become a prominent method of AI implementation in medical classification tasks. Grading Diabetic Retinopathy (DR) has been at the forefront of the development of AI for ophthalmology. However, major obstacles remain in the generalization of these CNNs onto real-world DR screening programs. We believe these difficulties are due to use of 1) small training datasets (<5,000 images), 2) private and 'curated' repositories, 3) locally implemented CNN implementation methods, while 4) relying on measured Area Under the Curve (AUC) as the sole measure of CNN performance. To address these issues, the public EyePACS Kaggle Diabetic Retinopathy dataset was uploaded onto Microsoft Azure™ cloud platform. Two CNNs were trained; 1 a "Quality Assurance", and 2. a "Classifier". The Diabetic Retinopathy classifier CNN (DRCNN) performance was then tested both on 'un-curated' as well as the 'curated' test set created by the "Quality Assessment" CNN model. Finally, the sensitivity of the DRCNNs was boosted using two post-training techniques. Our DRCNN proved to be robust, as its performance was similar on 'curated' and 'un-curated' test sets. The implementation of 'cascading thresholds' and 'max margin' techniques led to significant improvements in the DRCNN's sensitivity, while also enhancing the specificity of other grades. OPEN ACCESSCitation: Xie L, Yang S, Squirrell D, Vaghefi E (2020) Towards implementation of AI in New Zealand national diabetic screening program: Cloud-based, robust, and bespoke. PLoS ONE 15 (4): e0225015. https://doi.org/10.

show abstract

Section: Plos Onementioning

confidence: 99%

Towards implementation of AI in New Zealand national diabetic screening program: Cloud-based, robust, and bespoke

et al. 2020

View full text Add to dashboard Cite

show abstract

“…So, there is a huge spotlight on tomography and radiology fields. Last year a review on deep learning and ophthalmology 22 and, also last year, a review on deep learning and radiotherapy 19 were carried out.…”

Section: Deep Learningmentioning

confidence: 99%

An overview of machine learning in health related areas: pitfalls and opportunities

Vitorasso

Vitorasso²

2020

BJHR

View full text Add to dashboard Cite

Machine learning techniques are on the spotlight in current scientific literature and these methods are gaining prominence in the health field. However, there are a few considerations that must be taken before conducting a study with machine learning techniques. This paper aims to provide an overview of machine learning methods applied to studies of health related areas. Additionally, this article will discuss important points about data preparation that may influence on the prediction outcome; comparison with statistical analysis; and potential applications. A literature search was carried out, using IEEE xplore and Pubmed, of publications from the last 10 years. Undoubtedly machine learning is becoming more and more present in science. However, the unfamiliarity with this technology may hinder or jeopardize its application. As any scientific tool, machine learning presents positive points along with limitations and both aspects should be considered in every analysis. The researcher must select the most adequate method and consider all repercussions of data preparation on the predictive model. A special attention should be given towards distance based techniques. ML techniques are full with potential applications; however these methods did not replace classical statistical analysis and, yet, they will continue to be an important tool to in health areas.

show abstract

“…One of the most promising approaches for handling multiple parameters is automated machine learning methods that gained large popularity. These methods, and more specifically, the deep learning method, show encouraging results for diagnosis and prediction of disease progression [63,64].…”

Section: Glaucoma Diagnosismentioning

confidence: 99%

Comprehensive Glaucoma Imaging

Glidai¹,

Kahook²,

Noecker³

et al. 2020

Albert and Jakobiec's Principles and Practice of Ophthalmology

View full text Add to dashboard Cite

Advancements in ocular imaging over the past four decades have revolutionized glaucoma practice worldwide. As glaucomatous vision loss cannot be restored, early diagnosis, proper management, and timely intervention are essential to slow disease progression. The emergence of advanced computerized imaging modalities has shifted glaucoma assessment from being largely subjective to mostly objective. These modalities can document and accurately quantify the optic nerve head and the macula regions, which are affected by the disease. With various ocular imaging devices

show abstract

Deep learning in ophthalmology: a review

Cited by 112 publications

References 14 publications

Towards implementation of AI in New Zealand national diabetic screening program: Cloud-based, robust, and bespoke

Towards implementation of AI in New Zealand national diabetic screening program: Cloud-based, robust, and bespoke

An overview of machine learning in health related areas: pitfalls and opportunities

Comprehensive Glaucoma Imaging

Contact Info

Product

Resources

About