{"id":"W4387211423","doi":"10.1007/978-3-031-43990-2_63","title":"Lesion-Aware Contrastive Learning for Diabetic Retinopathy Diagnosis","year":2023,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; False positive paradox; Artificial intelligence; Discriminative model; False positives and false negatives; Diabetic retinopathy; Lesion; Deep learning; Pattern recognition (psychology); Grading (engineering); Machine learning; Supervised learning; Fundus (uterus); Feature learning; Medicine; Radiology; Pathology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001835613,0.0005002235,0.0003226231,0.0004170223,0.00009308378,0.0004631584,0.0006783077,0.0004835897,0.004363759],"category_scores_gemma":[0.0004838728,0.000135185,0.0003590904,0.000345203,0.0001138806,0.0004213949,0.000384549,0.0005868801,0.001199957],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001923691,"about_ca_system_score_gemma":0.0002453063,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001256726,"about_ca_topic_score_gemma":0.002323831,"domain_scores_codex":[0.9999286,0.000008750152,0.000002989354,0.00002029541,0.00002784003,0.00001148833],"domain_scores_gemma":[0.999871,0.00006762045,0.000009326164,0.00001310105,0.00003042775,0.000008525861],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001266345,0.0001036777,0.0005167211,0.0001314013,0.00003271607,0.0001311657,0.0000191804,0.01580655,0.04423946,0.002467771,0.006146916,0.9302778],"study_design_scores_gemma":[0.00002151125,0.0002815235,0.003130757,0.00008490948,0.0001173908,0.001737645,0.00004133815,0.8645638,0.1002643,0.0103731,0.01934745,0.00003628066],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04122721,0.008363578,0.9342864,0.0003989214,0.0002800351,0.00007920621,0.0004707894,0.002293285,0.01260061],"genre_scores_gemma":[0.396513,0.008978789,0.5654603,0.0005139542,0.0003539766,0.000074282,0.001204512,0.0002982857,0.02660282],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004363759,"threshold_uncertainty_score":0.01459819,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02669811786685074,"score_gpt":0.2892243224098505,"score_spread":0.2625262045429997,"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."}}