Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Apr 1;43(4):419-424.
doi: 10.1097/ICO.0000000000003318. Epub 2023 May 31.

Automatic Classification of Slit-Lamp Photographs by Imaging Illumination

Ming-Chen Lu  1 , Callie Deng  1 , Miles F Greenwald  1 , Sina Farsiu  2   3 , N Venkatesh Prajna  4 , Nambi Nallasamy  1   5 , Mercy Pawar  1 , Jenna N Hart  1 , Sumithra Sr  4 , Prabhleen Kochar  4 , Suvitha Selvaraj  4 , Harry Levine  6 , Guillermo Amescua  6 , Paula A Sepulveda-Beltran  6 , Leslie M Niziol  1 , Maria A Woodward  1   7 ; and the AQUA Study Team
Affiliations

Automatic Classification of Slit-Lamp Photographs by Imaging Illumination

Ming-Chen Lu et al. Cornea. .

Abstract

Purpose: The aim of this study was to facilitate deep learning systems in image annotations for diagnosing keratitis type by developing an automated algorithm to classify slit-lamp photographs (SLPs) based on illumination technique.

Methods: SLPs were collected from patients with corneal ulcer at Kellogg Eye Center, Bascom Palmer Eye Institute, and Aravind Eye Care Systems. Illumination techniques were slit beam, diffuse white light, diffuse blue light with fluorescein, and sclerotic scatter (ScS). Images were manually labeled for illumination and randomly split into training, validation, and testing data sets (70%:15%:15%). Classification algorithms including MobileNetV2, ResNet50, LeNet, AlexNet, multilayer perceptron, and k-nearest neighborhood were trained to distinguish 4 type of illumination techniques. The algorithm performances on the test data set were evaluated with 95% confidence intervals (CIs) for accuracy, F1 score, and area under the receiver operator characteristics curve (AUC-ROC), overall and by class (one-vs-rest).

Results: A total of 12,132 images from 409 patients were analyzed, including 41.8% (n = 5069) slit-beam photographs, 21.2% (2571) diffuse white light, 19.5% (2364) diffuse blue light, and 17.5% (2128) ScS. MobileNetV2 achieved the highest overall F1 score of 97.95% (CI, 97.94%-97.97%), AUC-ROC of 99.83% (99.72%-99.9%), and accuracy of 98.98% (98.97%-98.98%). The F1 scores for slit beam, diffuse white light, diffuse blue light, and ScS were 97.82% (97.80%-97.84%), 96.62% (96.58%-96.66%), 99.88% (99.87%-99.89%), and 97.59% (97.55%-97.62%), respectively. Slit beam and ScS were the 2 most frequently misclassified illumination.

Conclusions: MobileNetV2 accurately labeled illumination of SLPs using a large data set of corneal images. Effective, automatic classification of SLPs is key to integrating deep learning systems for clinical decision support into practice workflows.

PubMed Disclaimer

Conflict of interest statement

Funding for this research was provided by the National Eye Institute (R01EY031033, M.A.W), (P30 EY005722, S.F.), and a Research to Prevent Blindness Career Advancement Award (M.A.W). For the remaining authors no disclosures were declared. The funding support played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Figures

Figure 1.
Figure 1.
Slit lamp photograph illumination type, including: A. Diffuse blue light. B. Slit beam. C. Sclerotic scatter. D. Diffuse white light.
Figure 2.
Figure 2.
Data pipeline for classifying slit lamp photos by image modalities.
Figure 3.
Figure 3.
Samples of misclassification images by MobilNetV2, including: A. The true label was slit beam but was predicted as sclerotic scatter. B. The true label was slit beam but was predicted as diffuse white light. C. The true label was sclerotic scatter but was predicted as diffuse white light.
See this image and copyright information in PMC

References

    1. Flaxman SR, Bourne RRA, Resnikoff S, et al. Global causes of blindness and distance vision impairment 1990–2020: a systematic review and meta-analysis. Lancet Glob Health. 2017;5(12):e1221–e1234. - PubMed
    1. Whitcher JP, Srinivasan M, Upadhyay MP. Corneal blindness: a global perspective. Bull World Health Organ. 2001;79(3):214–221. - PMC - PubMed
    1. GBD 2019 Blindness and Vision Impairment Collaborators, Vision Loss Expert Group of the Global Burden of Disease Study. Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: the Right to Sight: an analysis for the Global Burden of Disease Study. Lancet Glob Health. 2021;9(2):e144–e160. - PMC - PubMed
    1. Collier SA, Gronostaj MP, MacGurn AK, et al. Estimated burden of keratitis--United States, 2010. MMWR Morb Mortal Wkly Rep. 2014;63(45):1027–1030. - PMC - PubMed
    1. Austin A, Lietman T, Rose-Nussbaumer J. Update on the Management of Infectious Keratitis. Ophthalmology. 2017;124(11):1678–1689. - PMC - PubMed