Evaluation of the corneal topography based on deep learning
PurposeThe current study designed a unique type of corneal topography evaluation method based on deep learning and traditional image processing algorithms. The type of corneal topography of patients was evaluated through the segmentation of important medical zones and the calculation of relevant med...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-01-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2023.1264659/full |
| _version_ | 1827391145176465408 |
|---|---|
| author | Shuai Xu Xiaoyan Yang Xiaoyan Yang Xiaoyan Yang Xiaoyan Yang Shuxian Zhang Shuxian Zhang Shuxian Zhang Shuxian Zhang Xuan Zheng Fang Zheng Yin Liu Hanyu Zhang Lihua Li Lihua Li Lihua Li Lihua Li Qing Ye |
| author_facet | Shuai Xu Xiaoyan Yang Xiaoyan Yang Xiaoyan Yang Xiaoyan Yang Shuxian Zhang Shuxian Zhang Shuxian Zhang Shuxian Zhang Xuan Zheng Fang Zheng Yin Liu Hanyu Zhang Lihua Li Lihua Li Lihua Li Lihua Li Qing Ye |
| author_sort | Shuai Xu |
| collection | DOAJ |
| description | PurposeThe current study designed a unique type of corneal topography evaluation method based on deep learning and traditional image processing algorithms. The type of corneal topography of patients was evaluated through the segmentation of important medical zones and the calculation of relevant medical indicators of orthokeratology (OK) lenses.MethodsThe clinical data of 1,302 myopic subjects was collected retrospectively. A series of neural network-based U-Net was used to segment the pupil and the treatment zone in the corneal topography, and the decentration, effective defocusing contact range, and other indicators were calculated according to the image processing algorithm. The type of corneal topography was evaluated according to the evaluation criteria given by the optometrist. Finally, the method described in this article was used to evaluate the type of corneal topography and compare it with the type classified by the optometrist.ResultsWhen the important medical zones in the corneal topography were segmented, the precision and recall of the treatment zone reached 0.9587 and 0.9459, respectively, and the precision and recall of the pupil reached 0.9771 and 0.9712. Finally, the method described in this article was used to evaluate the type of corneal topography. When the reviewed findings based on deep learning and image processing algorithms were compared to the type of corneal topography marked by the professional optometrist, they demonstrated high accuracy with more than 98%.ConclusionThe current study provided an effective and accurate deep learning algorithm to evaluate the type of corneal topography. The deep learning algorithm played an auxiliary role in the OK lens fitting, which could help optometrists select the parameters of OK lenses effectively. |
| first_indexed | 2024-03-08T17:06:28Z |
| format | Article |
| id | doaj.art-fab7fc225d00439895c17c816ed6d797 |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-03-08T17:06:28Z |
| publishDate | 2024-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-fab7fc225d00439895c17c816ed6d7972024-01-04T05:02:17ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2024-01-011010.3389/fmed.2023.12646591264659Evaluation of the corneal topography based on deep learningShuai Xu0Xiaoyan Yang1Xiaoyan Yang2Xiaoyan Yang3Xiaoyan Yang4Shuxian Zhang5Shuxian Zhang6Shuxian Zhang7Shuxian Zhang8Xuan Zheng9Fang Zheng10Yin Liu11Hanyu Zhang12Lihua Li13Lihua Li14Lihua Li15Lihua Li16Qing Ye17Key Laboratory of Weak-Light Nonlinear Photonics, Ministry of Education, School of Physics and TEDA Applied Physics, Nankai University, Tianjin, ChinaTianjin Eye Hospital, Tianjin, ChinaTianjin Key Lab of Ophthalmology and Visual Science, Tianjin, ChinaNankai University Affiliated Eye Hospital, Tianjin, ChinaEye Hospital Optometric Center, Tianjin, ChinaTianjin Eye Hospital, Tianjin, ChinaTianjin Key Lab of Ophthalmology and Visual Science, Tianjin, ChinaNankai University Affiliated Eye Hospital, Tianjin, ChinaEye Hospital Optometric Center, Tianjin, ChinaKey Laboratory of Weak-Light Nonlinear Photonics, Ministry of Education, School of Physics and TEDA Applied Physics, Nankai University, Tianjin, ChinaKey Laboratory of Weak-Light Nonlinear Photonics, Ministry of Education, School of Physics and TEDA Applied Physics, Nankai University, Tianjin, ChinaSchool of Medicine, Nankai University, Tianjin, ChinaSchool of Medicine, Nankai University, Tianjin, ChinaTianjin Eye Hospital, Tianjin, ChinaTianjin Key Lab of Ophthalmology and Visual Science, Tianjin, ChinaNankai University Affiliated Eye Hospital, Tianjin, ChinaEye Hospital Optometric Center, Tianjin, ChinaKey Laboratory of Weak-Light Nonlinear Photonics, Ministry of Education, School of Physics and TEDA Applied Physics, Nankai University, Tianjin, ChinaPurposeThe current study designed a unique type of corneal topography evaluation method based on deep learning and traditional image processing algorithms. The type of corneal topography of patients was evaluated through the segmentation of important medical zones and the calculation of relevant medical indicators of orthokeratology (OK) lenses.MethodsThe clinical data of 1,302 myopic subjects was collected retrospectively. A series of neural network-based U-Net was used to segment the pupil and the treatment zone in the corneal topography, and the decentration, effective defocusing contact range, and other indicators were calculated according to the image processing algorithm. The type of corneal topography was evaluated according to the evaluation criteria given by the optometrist. Finally, the method described in this article was used to evaluate the type of corneal topography and compare it with the type classified by the optometrist.ResultsWhen the important medical zones in the corneal topography were segmented, the precision and recall of the treatment zone reached 0.9587 and 0.9459, respectively, and the precision and recall of the pupil reached 0.9771 and 0.9712. Finally, the method described in this article was used to evaluate the type of corneal topography. When the reviewed findings based on deep learning and image processing algorithms were compared to the type of corneal topography marked by the professional optometrist, they demonstrated high accuracy with more than 98%.ConclusionThe current study provided an effective and accurate deep learning algorithm to evaluate the type of corneal topography. The deep learning algorithm played an auxiliary role in the OK lens fitting, which could help optometrists select the parameters of OK lenses effectively.https://www.frontiersin.org/articles/10.3389/fmed.2023.1264659/fulldeep learningimage processingcorneal topographyorthokeratology lenstreatment zone |
| spellingShingle | Shuai Xu Xiaoyan Yang Xiaoyan Yang Xiaoyan Yang Xiaoyan Yang Shuxian Zhang Shuxian Zhang Shuxian Zhang Shuxian Zhang Xuan Zheng Fang Zheng Yin Liu Hanyu Zhang Lihua Li Lihua Li Lihua Li Lihua Li Qing Ye Evaluation of the corneal topography based on deep learning Frontiers in Medicine deep learning image processing corneal topography orthokeratology lens treatment zone |
| title | Evaluation of the corneal topography based on deep learning |
| title_full | Evaluation of the corneal topography based on deep learning |
| title_fullStr | Evaluation of the corneal topography based on deep learning |
| title_full_unstemmed | Evaluation of the corneal topography based on deep learning |
| title_short | Evaluation of the corneal topography based on deep learning |
| title_sort | evaluation of the corneal topography based on deep learning |
| topic | deep learning image processing corneal topography orthokeratology lens treatment zone |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2023.1264659/full |
| work_keys_str_mv | AT shuaixu evaluationofthecornealtopographybasedondeeplearning AT xiaoyanyang evaluationofthecornealtopographybasedondeeplearning AT xiaoyanyang evaluationofthecornealtopographybasedondeeplearning AT xiaoyanyang evaluationofthecornealtopographybasedondeeplearning AT xiaoyanyang evaluationofthecornealtopographybasedondeeplearning AT shuxianzhang evaluationofthecornealtopographybasedondeeplearning AT shuxianzhang evaluationofthecornealtopographybasedondeeplearning AT shuxianzhang evaluationofthecornealtopographybasedondeeplearning AT shuxianzhang evaluationofthecornealtopographybasedondeeplearning AT xuanzheng evaluationofthecornealtopographybasedondeeplearning AT fangzheng evaluationofthecornealtopographybasedondeeplearning AT yinliu evaluationofthecornealtopographybasedondeeplearning AT hanyuzhang evaluationofthecornealtopographybasedondeeplearning AT lihuali evaluationofthecornealtopographybasedondeeplearning AT lihuali evaluationofthecornealtopographybasedondeeplearning AT lihuali evaluationofthecornealtopographybasedondeeplearning AT lihuali evaluationofthecornealtopographybasedondeeplearning AT qingye evaluationofthecornealtopographybasedondeeplearning |