Keratoconus detection of changes using deep learning of colour-coded maps

Chen, Xu, Zhao, Jiaxin, Iselin, Katja C, Borroni, Davide ORCID: 0000-0001-6952-5647, Romano, Davide, Gokul, Akilesh, McGhee, Charles NJ, Zhao, Yitian, Sedaghat, Mohammad-Reza, Momeni-Moghaddam, Hamed
et al (show 4 more authors) (2021) Keratoconus detection of changes using deep learning of colour-coded maps. BMJ Open Ophthalmology, 6 (1). e000824-e000824.

Access the full-text of this item by clicking on the Open Access link.


<jats:sec><jats:title>Objective</jats:title><jats:p>To evaluate the accuracy of convolutional neural networks technique (CNN) in detecting keratoconus using colour-coded corneal maps obtained by a Scheimpflug camera.</jats:p></jats:sec><jats:sec><jats:title>Design</jats:title><jats:p>Multicentre retrospective study.</jats:p></jats:sec><jats:sec><jats:title>Methods and analysis</jats:title><jats:p>We included the images of keratoconic and healthy volunteers’ eyes provided by three centres: Royal Liverpool University Hospital (Liverpool, UK), Sedaghat Eye Clinic (Mashhad, Iran) and The New Zealand National Eye Center (New Zealand). Corneal tomography scans were used to train and test CNN models, which included healthy controls. Keratoconic scans were classified according to the Amsler-Krumeich classification. Keratoconic scans from Iran were used as an independent testing set. Four maps were considered for each scan: axial map, anterior and posterior elevation map, and pachymetry map.</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>A CNN model detected keratoconus versus health eyes with an accuracy of 0.9785 on the testing set, considering all four maps concatenated. Considering each map independently, the accuracy was 0.9283 for axial map, 0.9642 for thickness map, 0.9642 for the front elevation map and 0.9749 for the back elevation map. The accuracy of models in recognising between healthy controls and stage 1 was 0.90, between stages 1 and 2 was 0.9032, and between stages 2 and 3 was 0.8537 using the concatenated map.</jats:p></jats:sec><jats:sec><jats:title>Conclusion</jats:title><jats:p>CNN provides excellent detection performance for keratoconus and accurately grades different severities of disease using the colour-coded maps obtained by the Scheimpflug camera. CNN has the potential to be further developed, validated and adopted for screening and management of keratoconus.</jats:p></jats:sec>

Item Type: Article
Uncontrolled Keywords: cornea, imaging
Divisions: Faculty of Health and Life Sciences
Faculty of Health and Life Sciences > Clinical Directorate
Faculty of Health and Life Sciences > Institute of Life Courses and Medical Sciences
Depositing User: Symplectic Admin
Date Deposited: 02 Aug 2021 07:41
Last Modified: 18 Jan 2023 21:34
DOI: 10.1136/bmjophth-2021-000824
Open Access URL:
Related URLs: