{"doi":"10.1016/j.xops.2023.100389","title":"An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging","abstract":"PurposeTo develop an objective glaucoma damage severity classification system based on optical coherence tomography (OCT)-derived retinal nerve fiber layer (RNFL) thickness measurements.DesignAlgorithm development for RNFL damage severity classification based on multicenter OCT data.Subjects and ParticipantsA 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.MethodsWe 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 MeasuresAccuracy, area under the receiver operating characteristic curve (AUC), and confusion matrix.ResultsThe k-means clustering discovered four clusters with 1382, 1613, 1727, and 1839 samples with mean global RNFL thickness of 58.3 μm (±8.9: Standard Deviation), 78.9 μm (±6.7), 87.7 μm (±8.2), and 101.5 μm (±7.9). The Bayes’ minimum error classifier identified optimal global RNFL values of >95 μm, 86 – 95 μm, and 70 – 85 μm and <70 μm for discriminating normal eyes and eyes at the early, moderate, and advanced stages of RNFL thickness loss. About 4% of normal eyes and 98% of eyes with advanced RNFL loss had either global or at least one quadrant RNFL thickness outside of normal limits provided by the OCT instrument.ConclusionsUnsupervised 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 pre-assumption 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.","journal":"Ophthalmology Science","year":2023,"id":354122,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":7,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.5384,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2023-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":391791,"name":"Xiaoqin Huang","orcid":null,"position":1,"is_corresponding":false},{"id":1101779,"name":"Asma Poursoroush","orcid":null,"position":2,"is_corresponding":false},{"id":1101780,"name":"Julek Majoor","orcid":null,"position":3,"is_corresponding":false},{"id":709221,"name":"Hans G. Lemij","orcid":"0000-0001-8508-8985","position":4,"is_corresponding":false},{"id":709220,"name":"Koenraad A. Vermeer","orcid":"0000-0002-4038-3945","position":5,"is_corresponding":false},{"id":259255,"name":"Tobias Elze","orcid":"0000-0002-2032-0496","position":6,"is_corresponding":false},{"id":342311,"name":"Mengyu Wang","orcid":"0000-0002-7188-7126","position":7,"is_corresponding":false},{"id":315555,"name":"Kouros Nouri‐Mahdavi","orcid":"0000-0001-9403-8904","position":8,"is_corresponding":false},{"id":316297,"name":"Vahid Mohammadzadeh","orcid":null,"position":9,"is_corresponding":false},{"id":894327,"name":"Paolo Brusini","orcid":"0000-0003-0855-5775","position":10,"is_corresponding":false},{"id":307582,"name":"Chris A. Johnson","orcid":"0000-0002-0090-630X","position":11,"is_corresponding":false},{"id":281320,"name":"Siamak Yousefi","orcid":"0000-0001-8633-5730","position":0,"is_corresponding":true}],"reference_count":45,"raw_metadata":null,"created_at":"2026-07-19T01:13:06.959762Z","pmid":"37868793","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}