CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning
Deep learning has dramatically improved object recognition, speech recognition, medical image analysis and many other fields. Optical coherence tomography (OCT) has become a standard of care imaging modality for ophthalmology. We asked whether deep learning could be used to segment cornea OCT images...
| Main Authors: | , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/105553 http://hdl.handle.net/10220/47814 |
| _version_ | 1826127017064529920 |
|---|---|
| author | Santos, Valentin Aranha dos Schmetterer, Leopold Stegmann, Hannes Pfister, Martin Messner, Alina Schmidinger, Gerald Garhofer, Gerhard Werkmeister, René M. |
| author2 | Lee Kong Chian School of Medicine (LKCMedicine) |
| author_facet | Lee Kong Chian School of Medicine (LKCMedicine) Santos, Valentin Aranha dos Schmetterer, Leopold Stegmann, Hannes Pfister, Martin Messner, Alina Schmidinger, Gerald Garhofer, Gerhard Werkmeister, René M. |
| author_sort | Santos, Valentin Aranha dos |
| collection | NTU |
| description | Deep learning has dramatically improved object recognition, speech recognition, medical image analysis and many other fields. Optical coherence tomography (OCT) has become a standard of care imaging modality for ophthalmology. We asked whether deep learning could be used to segment cornea OCT images. Using a custom-built ultrahigh-resolution OCT system, we scanned 72 healthy eyes and 70 keratoconic eyes. In total, 20,160 images were labeled and used for the training in a supervised learning approach. A custom neural network architecture called CorneaNet was designed and trained. Our results show that CorneaNet is able to segment both healthy and keratoconus images with high accuracy (validation accuracy: 99.56%). Thickness maps of the three main corneal layers (epithelium, Bowman’s layer and stroma) were generated both in healthy subjects and subjects suffering from keratoconus. CorneaNet is more than 50 times faster than our previous algorithm. Our results show that deep learning algorithm scan be used for OCT image segmentation and could be applied in various clinical settings. In particular, CorneaNet could be used for early detection of keratoconus and more generally to study other diseases altering corneal morphology. |
| first_indexed | 2024-10-01T07:02:12Z |
| format | Journal Article |
| id | ntu-10356/105553 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:02:12Z |
| publishDate | 2019 |
| record_format | dspace |
| spelling | ntu-10356/1055532020-11-01T05:26:42Z CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning Santos, Valentin Aranha dos Schmetterer, Leopold Stegmann, Hannes Pfister, Martin Messner, Alina Schmidinger, Gerald Garhofer, Gerhard Werkmeister, René M. Lee Kong Chian School of Medicine (LKCMedicine) Deep Learning Algorithms DRNTU::Science::Medicine Optical Coherence Tomography Deep learning has dramatically improved object recognition, speech recognition, medical image analysis and many other fields. Optical coherence tomography (OCT) has become a standard of care imaging modality for ophthalmology. We asked whether deep learning could be used to segment cornea OCT images. Using a custom-built ultrahigh-resolution OCT system, we scanned 72 healthy eyes and 70 keratoconic eyes. In total, 20,160 images were labeled and used for the training in a supervised learning approach. A custom neural network architecture called CorneaNet was designed and trained. Our results show that CorneaNet is able to segment both healthy and keratoconus images with high accuracy (validation accuracy: 99.56%). Thickness maps of the three main corneal layers (epithelium, Bowman’s layer and stroma) were generated both in healthy subjects and subjects suffering from keratoconus. CorneaNet is more than 50 times faster than our previous algorithm. Our results show that deep learning algorithm scan be used for OCT image segmentation and could be applied in various clinical settings. In particular, CorneaNet could be used for early detection of keratoconus and more generally to study other diseases altering corneal morphology. Published version 2019-03-14T07:03:12Z 2019-12-06T21:53:30Z 2019-03-14T07:03:12Z 2019-12-06T21:53:30Z 2019 Journal Article Santos, V. A. d., Schmetterer, L., Stegmann, H., Pfister, M., Messner, A., Schmidinger, G., . . . Werkmeister, R. M. (2019). CorneaNet: fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning. Biomedical Optics Express, 10(2), 622-641. doi:10.1364/BOE.10.000622 https://hdl.handle.net/10356/105553 http://hdl.handle.net/10220/47814 10.1364/BOE.10.000622 en Biomedical Optics Express © 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement. Users may use, reuse, and build upon the article, or use the article for text or data mining, so long as such uses are for non-commercial purposes and appropriate attribution is maintained. All other rights are reserved. 20 p. application/pdf |
| spellingShingle | Deep Learning Algorithms DRNTU::Science::Medicine Optical Coherence Tomography Santos, Valentin Aranha dos Schmetterer, Leopold Stegmann, Hannes Pfister, Martin Messner, Alina Schmidinger, Gerald Garhofer, Gerhard Werkmeister, René M. CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning |
| title | CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning |
| title_full | CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning |
| title_fullStr | CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning |
| title_full_unstemmed | CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning |
| title_short | CorneaNet : fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning |
| title_sort | corneanet fast segmentation of cornea oct scans of healthy and keratoconic eyes using deep learning |
| topic | Deep Learning Algorithms DRNTU::Science::Medicine Optical Coherence Tomography |
| url | https://hdl.handle.net/10356/105553 http://hdl.handle.net/10220/47814 |
| work_keys_str_mv | AT santosvalentinaranhados corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning AT schmettererleopold corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning AT stegmannhannes corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning AT pfistermartin corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning AT messneralina corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning AT schmidingergerald corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning AT garhofergerhard corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning AT werkmeisterrenem corneanetfastsegmentationofcorneaoctscansofhealthyandkeratoconiceyesusingdeeplearning |