{"id":"W4283361787","doi":"10.1007/s10633-022-09879-7","title":"MERCI: a machine learning approach to identifying hydroxychloroquine retinopathy using mfERG","year":2022,"lang":"en","type":"article","venue":"Documenta Ophthalmologica","topic":"Drug-Induced Ocular Toxicity","field":"Medicine","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"Kensington Health; York University; University of Toronto","funders":"Mitacs","keywords":"Hydroxychloroquine; Medicine; Retinopathy; Retinal; Ophthalmology; Clinical trial; Optometry; Artificial intelligence; Disease; Internal medicine; Computer science; Coronavirus disease 2019 (COVID-19)","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007178707,0.0009010775,0.001083296,0.001625881,0.0004352495,0.0008472306,0.001093731,0.001184515,0.002673226],"category_scores_gemma":[0.001138181,0.0003105463,0.0008992626,0.0005372004,0.0001272847,0.0004993044,0.0007685115,0.000897,0.001345065],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003612808,"about_ca_system_score_gemma":0.0005551957,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004953095,"about_ca_topic_score_gemma":0.007679152,"domain_scores_codex":[0.9997788,0.0000418519,0.00001266555,0.00005923057,0.0000752162,0.00003224345],"domain_scores_gemma":[0.9997117,0.000112615,0.00003421375,0.00003209147,0.00008831549,0.00002110555],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008282135,0.0006413768,0.01572857,0.0001548058,0.0004798198,0.0002733868,0.00005045769,0.04394919,0.02842386,0.001145683,0.01944086,0.8888838],"study_design_scores_gemma":[0.00004940797,0.0001973343,0.009569166,0.00001886848,0.00008174527,0.0002617221,0.00002699414,0.9750954,0.009978839,0.001277498,0.003400466,0.00004267829],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1024582,0.002093403,0.8606551,0.0008675735,0.0002914456,0.0005051653,0.004478324,0.02288818,0.005762549],"genre_scores_gemma":[0.4129225,0.0007309875,0.5679855,0.0004180368,0.0003286727,0.0004193871,0.004424466,0.0003753193,0.01239525],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004953095,"threshold_uncertainty_score":0.009848535,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07134996811524971,"score_gpt":0.3306422252678251,"score_spread":0.2592922571525753,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}