{"id":"W4283800998","doi":"10.1609/aaai.v36i11.21452","title":"Evaluating Explainable AI on a Multi-Modal Medical Imaging Task: Can Existing Algorithms Fulfill Clinical Requirements?","year":2022,"lang":"en","type":"article","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":61,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; Simon Fraser University","funders":"BC Cancer Foundation; Compute Canada; Nvidia","keywords":"Computer science; Modality (human–computer interaction); Artificial intelligence; Leverage (statistics); Modal; Machine learning; Metric (unit); Feature (linguistics); Medical imaging; Data mining; Engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.009532185,0.0003610786,0.0004487962,0.0002838503,0.001686011,0.0003786586,0.002680806,0.00008894761,0.001142317],"category_scores_gemma":[0.002996735,0.0003621658,0.0002178698,0.001017512,0.000177219,0.0007901887,0.00267439,0.00119714,0.0001863335],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005524363,"about_ca_system_score_gemma":0.0007903617,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001663683,"about_ca_topic_score_gemma":0.0001625253,"domain_scores_codex":[0.9920766,0.001110951,0.001385839,0.001474827,0.00270225,0.001249525],"domain_scores_gemma":[0.9966084,0.0009778894,0.0003397887,0.001220895,0.000313827,0.0005392157],"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":[0.00007200803,0.001577637,0.004089578,0.00003373439,0.00007891975,0.001622128,0.00385466,0.008861851,0.00120659,0.08074598,0.008242092,0.8896148],"study_design_scores_gemma":[0.0006250128,0.0004664523,0.0001522275,0.00004123866,0.000009693686,0.00008462306,0.00223465,0.985373,0.001294434,0.002736992,0.006513797,0.0004679375],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03044361,0.0001239844,0.9392419,0.01523774,0.003240296,0.0008612267,0.00001052739,0.0009227057,0.009917974],"genre_scores_gemma":[0.8331841,0.000009324244,0.1449094,0.0178121,0.0005552571,0.0003772998,0.00001856482,0.00006826501,0.003065725],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9765111,"threshold_uncertainty_score":0.9998831,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2237399218700203,"score_gpt":0.4682253697246115,"score_spread":0.2444854478545912,"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."}}