Dataset Information

Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study.

ABSTRACT:

Objective

To evaluate the diagnostic accuracy of keratoconus using deep learning of the colour-coded maps measured with the swept-source anterior segment optical coherence tomography (AS-OCT).

Design

A diagnostic accuracy study.

Setting

A single-centre study.

Participants

A total of 304 keratoconic eyes (grade 1 (108 eyes), 2 (75 eyes), 3 (42 eyes) and 4 (79 eyes)) according to the Amsler-Krumeich classification, and 239 age-matched healthy eyes.

Main outcome measures

The diagnostic accuracy of keratoconus using deep learning of six colour-coded maps (anterior elevation, anterior curvature, posterior elevation, posterior curvature, total refractive power and pachymetry map).

Results

Deep learning of the arithmetical mean output data of these six maps showed an accuracy of 0.991 in discriminating between normal and keratoconic eyes. For single map analysis, posterior elevation map (0.993) showed the highest accuracy, followed by posterior curvature map (0.991), anterior elevation map (0.983), corneal pachymetry map (0.982), total refractive power map (0.978) and anterior curvature map (0.976), in discriminating between normal and keratoconic eyes. This deep learning also showed an accuracy of 0.874 in classifying the stage of the disease. Posterior curvature map (0.869) showed the highest accuracy, followed by corneal pachymetry map (0.845), anterior curvature map (0.836), total refractive power map (0.836), posterior elevation map (0.829) and anterior elevation map (0.820), in classifying the stage.

Conclusions

Deep learning using the colour-coded maps obtained by the AS-OCT effectively discriminates keratoconus from normal corneas, and furthermore classifies the grade of the disease. It is suggested that this will become an aid for improving the diagnostic accuracy of keratoconus in daily practice.

Clinical trial registration number

000034587.

SUBMITTER: Kamiya K

PROVIDER: S-EPMC6773416 | biostudies-literature | 2019 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study.

Kamiya Kazutaka K Ayatsuka Yuji Y Kato Yudai Y Fujimura Fusako F Takahashi Masahide M Shoji Nobuyuki N Mori Yosai Y Miyata Kazunori K

BMJ open 20190927 9

<h4>Objective</h4>To evaluate the diagnostic accuracy of keratoconus using deep learning of the colour-coded maps measured with the swept-source anterior segment optical coherence tomography (AS-OCT).<h4>Design</h4>A diagnostic accuracy study.<h4>Setting</h4>A single-centre study.<h4>Participants</h4>A total of 304 keratoconic eyes (grade 1 (108 eyes), 2 (75 eyes), 3 (42 eyes) and 4 (79 eyes)) according to the Amsler-Krumeich classification, and 239 age-matched healthy eyes.<h4>Main outcome meas ...[more]

PMID: 31562158

Similar Datasets

Project description:PurposeThis study aimed to evaluate and compare the discriminating ability of corneal elevation maps generated using a swept-source optical coherence tomography (SS-OCT) (SS-OCT ANTERION, Heidelberg Engineering, Heidelberg, Germany), which was estimated with different reference surfaces, to distinguish normal corneas from those with keratoconus and keratoconus suspect.MethodsA total of 126 eyes of patients, which comprised 43, 37, and 46 keratoconus, keratoconus suspects, and normal controls, respectively, were included in this study. The anterior and posterior elevations at the thinnest point under the best-fit sphere (BFS) and toric-ellipsoid (BFT), respectively, and other corneal parameters were measured using the SS-OCT. In addition, the receiver operating characteristic (ROC) curve analysis and cut-off value were calculated to evaluate the diagnostic ability of the corneal elevation values in differentiating keratoconus and keratoconus suspects from normal eyes.ResultsThe mean total keratometric and corneal elevation values were significantly higher in the keratoconus group than in the other groups. Pachymetric parameters exhibited the lowest values for keratoconus. In addition, ROC curve analyses showed a high accuracy of the thinnest point anterior and posterior BFT for both keratoconus and keratoconus suspects and normal controls (area under the ROC were 0.969 and 0.961, respectively). Furthermore, the optimal cut-off point of the posterior elevation at the thinnest point under BFT was 16.44 μm (sensitivity and specificity of 86% and 98%, respectively) for differentiating keratoconus from normal and keratoconus suspect eyes.ConclusionsThe elevation map using the BFS and BFT references measured with the anterior segment SS-OCT is considered an effective indicator for keratoconus diagnosis. Therefore, the anterior segment SS-OCT can effectively differentiate keratoconus from suspected keratoconus and normal corneas by measuring parameters such as posterior and anterior elevations, pachymetry, and keratometry.

Dataset Information

Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study.

Objective

Design

Setting

Participants

Main outcome measures

Results

Conclusions

Clinical trial registration number

Publications

Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets