{"id":"W4294975594","doi":"10.1109/iri54793.2022.00046","title":"Using SHAP Analysis to Detect Areas Contributing to Diabetic Retinopathy Detection","year":2022,"lang":"en","type":"article","venue":"","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Machine learning; Computer science; Artificial intelligence; Context (archaeology); Reliability (semiconductor); Transfer of learning; Diabetic retinopathy; Population; Software deployment; Binary classification; Predictive modelling; Field (mathematics); Blindness; Deep learning; Support vector machine; Optometry; Diabetes mellitus; Medicine; Environmental health; Mathematics; Geography","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008753564,0.0001475096,0.0004542686,0.0008520628,0.0004715043,0.00004637372,0.00009922212,0.00002419226,0.0006628312],"category_scores_gemma":[0.0004616709,0.0001378548,0.0003138323,0.00362866,0.00001260867,0.00003147337,0.0001970753,0.0002109326,0.00003673049],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003117225,"about_ca_system_score_gemma":0.00003321544,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006799899,"about_ca_topic_score_gemma":0.00005394118,"domain_scores_codex":[0.9982146,0.0001287457,0.0003184782,0.0004217887,0.0004493985,0.000466989],"domain_scores_gemma":[0.9990727,0.00006920082,0.00007526443,0.0003368748,0.0001509421,0.0002950167],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004801883,0.0001327197,0.3323143,0.00003935172,0.002232485,0.0002180621,0.0007154621,0.03492867,0.567312,0.00001375652,0.0001809164,0.0614321],"study_design_scores_gemma":[0.002135193,0.00200629,0.2238028,0.0001118343,0.02028659,0.0003182454,0.004651574,0.4847012,0.254535,0.0001312331,0.005966049,0.001354111],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8841653,0.00005203942,0.1134758,0.00111638,0.00006665487,0.0002134974,0.000007196034,0.000114799,0.0007883458],"genre_scores_gemma":[0.9933735,8.606831e-7,0.003865147,0.001674118,0.00008762569,0.00003434345,0.00001223059,0.00001979235,0.0009323425],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4497725,"threshold_uncertainty_score":0.7257538,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02269491636037438,"score_gpt":0.304924706484282,"score_spread":0.2822297901239077,"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."}}