An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging
Purpose: To develop an objective glaucoma damage severity classification system based on OCT-derived retinal nerve fiber layer (RNFL) thickness measurements. Design: Algorithm development for RNFL damage severity classification based on multicenter OCT data. Subjects and Participants: A total of 656...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-03-01
|
| Series: | Ophthalmology Science |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666914523001215 |
| _version_ | 1827316908808994816 |
|---|---|
| author | Siamak Yousefi, PhD Xiaoqin Huang, PhD Asma Poursoroush, MSc Julek Majoor, PhD Hans Lemij, MD Koen Vermeer, PhD Tobias Elze, PhD Mengyu Wang, PhD Kouros Nouri-Mahdavi, MD, MPH Vahid Mohammadzadeh, MD Paolo Brusini, MD Chris Johnson, PhD |
| author_facet | Siamak Yousefi, PhD Xiaoqin Huang, PhD Asma Poursoroush, MSc Julek Majoor, PhD Hans Lemij, MD Koen Vermeer, PhD Tobias Elze, PhD Mengyu Wang, PhD Kouros Nouri-Mahdavi, MD, MPH Vahid Mohammadzadeh, MD Paolo Brusini, MD Chris Johnson, PhD |
| author_sort | Siamak Yousefi, PhD |
| collection | DOAJ |
| description | Purpose: To develop an objective glaucoma damage severity classification system based on OCT-derived retinal nerve fiber layer (RNFL) thickness measurements. Design: Algorithm development for RNFL damage severity classification based on multicenter OCT data. Subjects and Participants: A total of 6561 circumpapillary RNFL profiles from 2269 eyes of 1171 subjects to develop models, and 2505 RNFL profiles from 1099 eyes of 900 subjects to validate models. Methods: We developed an unsupervised k-means model to identify clusters of eyes with similar RNFL thickness profiles. We annotated the clusters based on their respective global RNFL thickness. We computed the optimal global RNFL thickness thresholds that discriminated different severity levels based on Bayes’ minimum error principle. We validated the proposed pipeline based on an independent validation dataset with 2505 RNFL profiles from 1099 eyes of 900 subjects. Main Outcome Measures: Accuracy, area under the receiver operating characteristic curve, and confusion matrix. Results: The k-means clustering discovered 4 clusters with 1382, 1613, 1727, and 1839 samples with mean (standard deviation) global RNFL thickness of 58.3 (8.9) μm, 78.9 (6.7) μm, 87.7 (8.2) μm, and 101.5 (7.9) μm. The Bayes’ minimum error classifier identified optimal global RNFL values of > 95 μm, 86 to 95 μm, 70 to 85 μm, and < 70 μm for discriminating normal eyes and eyes at the early, moderate, and advanced stages of RNFL thickness loss, respectively. About 4% of normal eyes and 98% of eyes with advanced RNFL loss had either global, or ≥ 1 quadrant, RNFL thickness outside of normal limits provided by the OCT instrument. Conclusions: Unsupervised machine learning discovered that the optimal RNFL thresholds for separating normal eyes and eyes with early, moderate, and advanced RNFL loss were 95 μm, 85 μm, and 70 μm, respectively. This RNFL loss classification system is unbiased as there was no preassumption or human expert intervention in the development process. Additionally, it is objective, easy to use, and consistent, which may augment glaucoma research and day-to-day clinical practice. Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article. |
| first_indexed | 2024-03-11T17:50:51Z |
| format | Article |
| id | doaj.art-49e40736a10f45428c8e079d36c0d471 |
| institution | Directory Open Access Journal |
| issn | 2666-9145 |
| language | English |
| last_indexed | 2024-04-24T23:22:36Z |
| publishDate | 2024-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ophthalmology Science |
| spelling | doaj.art-49e40736a10f45428c8e079d36c0d4712024-03-16T05:09:44ZengElsevierOphthalmology Science2666-91452024-03-0142100389An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity StagingSiamak Yousefi, PhD0Xiaoqin Huang, PhD1Asma Poursoroush, MSc2Julek Majoor, PhD3Hans Lemij, MD4Koen Vermeer, PhD5Tobias Elze, PhD6Mengyu Wang, PhD7Kouros Nouri-Mahdavi, MD, MPH8Vahid Mohammadzadeh, MD9Paolo Brusini, MD10Chris Johnson, PhD11Department of Ophthalmology, University of Tennessee Health Science Center, Memphis, Tennessee; Department of Genetics, Genomics, and Informatics, University of Tennessee Health Science Center, Memphis, Tennessee; Correspondence: Siamak Yousefi, PhD, University of Tennessee Health Science Center, 930 Madison Ave., Suite 726, Memphis, TN 38163.Department of Ophthalmology, University of Tennessee Health Science Center, Memphis, TennesseeDepartment of Ophthalmology, University of Tennessee Health Science Center, Memphis, TennesseeRotterdam Ophthalmic Institute, The Rotterdam Eye Hospital, Rotterdam, The NetherlandsRotterdam Ophthalmic Institute, The Rotterdam Eye Hospital, Rotterdam, The NetherlandsRotterdam Ophthalmic Institute, The Rotterdam Eye Hospital, Rotterdam, The NetherlandsSchepens Eye Research Institute of Massachusetts Eye and Ear, Harvard Medical School, Boston, MassachussettsSchepens Eye Research Institute of Massachusetts Eye and Ear, Harvard Medical School, Boston, MassachussettsDepartment of Ophthalmology, University of California Los Angeles, Los Angeles, CaliforniaDepartment of Ophthalmology, University of California Los Angeles, Los Angeles, CaliforniaDepartment of Ophthalmology, “Città di Udine” Health Center, Udine, ItalyDepartment of Ophthalmology & Visual Sciences, University of Iowa Hospitals and Clinics, Iowa City, IowaPurpose: To develop an objective glaucoma damage severity classification system based on OCT-derived retinal nerve fiber layer (RNFL) thickness measurements. Design: Algorithm development for RNFL damage severity classification based on multicenter OCT data. Subjects and Participants: A total of 6561 circumpapillary RNFL profiles from 2269 eyes of 1171 subjects to develop models, and 2505 RNFL profiles from 1099 eyes of 900 subjects to validate models. Methods: We developed an unsupervised k-means model to identify clusters of eyes with similar RNFL thickness profiles. We annotated the clusters based on their respective global RNFL thickness. We computed the optimal global RNFL thickness thresholds that discriminated different severity levels based on Bayes’ minimum error principle. We validated the proposed pipeline based on an independent validation dataset with 2505 RNFL profiles from 1099 eyes of 900 subjects. Main Outcome Measures: Accuracy, area under the receiver operating characteristic curve, and confusion matrix. Results: The k-means clustering discovered 4 clusters with 1382, 1613, 1727, and 1839 samples with mean (standard deviation) global RNFL thickness of 58.3 (8.9) μm, 78.9 (6.7) μm, 87.7 (8.2) μm, and 101.5 (7.9) μm. The Bayes’ minimum error classifier identified optimal global RNFL values of > 95 μm, 86 to 95 μm, 70 to 85 μm, and < 70 μm for discriminating normal eyes and eyes at the early, moderate, and advanced stages of RNFL thickness loss, respectively. About 4% of normal eyes and 98% of eyes with advanced RNFL loss had either global, or ≥ 1 quadrant, RNFL thickness outside of normal limits provided by the OCT instrument. Conclusions: Unsupervised machine learning discovered that the optimal RNFL thresholds for separating normal eyes and eyes with early, moderate, and advanced RNFL loss were 95 μm, 85 μm, and 70 μm, respectively. This RNFL loss classification system is unbiased as there was no preassumption or human expert intervention in the development process. Additionally, it is objective, easy to use, and consistent, which may augment glaucoma research and day-to-day clinical practice. Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.http://www.sciencedirect.com/science/article/pii/S2666914523001215Artificial intelligenceGlaucomaGlaucoma severity damageOptical coherence tomographyRetinal nerve fiber layer (RNFL)Staging |
| spellingShingle | Siamak Yousefi, PhD Xiaoqin Huang, PhD Asma Poursoroush, MSc Julek Majoor, PhD Hans Lemij, MD Koen Vermeer, PhD Tobias Elze, PhD Mengyu Wang, PhD Kouros Nouri-Mahdavi, MD, MPH Vahid Mohammadzadeh, MD Paolo Brusini, MD Chris Johnson, PhD An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging Ophthalmology Science Artificial intelligence Glaucoma Glaucoma severity damage Optical coherence tomography Retinal nerve fiber layer (RNFL) Staging |
| title | An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging |
| title_full | An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging |
| title_fullStr | An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging |
| title_full_unstemmed | An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging |
| title_short | An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging |
| title_sort | artificial intelligence enabled system for retinal nerve fiber layer thickness damage severity staging |
| topic | Artificial intelligence Glaucoma Glaucoma severity damage Optical coherence tomography Retinal nerve fiber layer (RNFL) Staging |
| url | http://www.sciencedirect.com/science/article/pii/S2666914523001215 |
| work_keys_str_mv | AT siamakyousefiphd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT xiaoqinhuangphd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT asmapoursoroushmsc anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT julekmajoorphd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT hanslemijmd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT koenvermeerphd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT tobiaselzephd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT mengyuwangphd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT kourosnourimahdavimdmph anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT vahidmohammadzadehmd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT paolobrusinimd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT chrisjohnsonphd anartificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT siamakyousefiphd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT xiaoqinhuangphd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT asmapoursoroushmsc artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT julekmajoorphd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT hanslemijmd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT koenvermeerphd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT tobiaselzephd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT mengyuwangphd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT kourosnourimahdavimdmph artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT vahidmohammadzadehmd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT paolobrusinimd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging AT chrisjohnsonphd artificialintelligenceenabledsystemforretinalnervefiberlayerthicknessdamageseveritystaging |