{"id":"W3215207525","doi":"10.1186/s13195-021-00924-2","title":"Improving 3D convolutional neural network comprehensibility via interactive visualization of relevance maps: evaluation in Alzheimer's disease.","year":2021,"lang":"en","type":"article","venue":"PUBLISSO (German National Library of Medicine)","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":65,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute on Aging; Canadian Institutes of Health Research; Deutsche Forschungsgemeinschaft","keywords":"Convolutional neural network; Relevance (law); Dementia; Computer science; Visualization; Discriminative model; Magnetic resonance imaging; Artificial intelligence; Atrophy; Clinical significance; Pattern recognition (psychology); Disease; Medicine; Pathology; Radiology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004493478,0.002538167,0.0008365593,0.001648235,0.0002848507,0.001346792,0.001577219,0.001628208,0.001441039],"category_scores_gemma":[0.01627264,0.0006376958,0.001535486,0.0005542094,0.0003985803,0.001244075,0.001352447,0.001234949,0.0003468149],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001192922,"about_ca_system_score_gemma":0.0006808145,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009068538,"about_ca_topic_score_gemma":0.007939409,"domain_scores_codex":[0.9991148,0.0003699632,0.00006661761,0.0002038052,0.0001731487,0.00007166342],"domain_scores_gemma":[0.9948835,0.003768055,0.0003071584,0.0003599974,0.0004940438,0.0001871661],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.003286512,0.0008387533,0.02460592,0.001489566,0.001556191,0.001132727,0.0006296397,0.5610827,0.02023858,0.001261318,0.01137474,0.3725034],"study_design_scores_gemma":[0.00008460702,0.0003256099,0.003599447,0.00005466783,0.0001275323,0.000222292,0.00003290282,0.9862148,0.007199948,0.001210468,0.0008939364,0.00003380872],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7641643,0.01126205,0.1973686,0.001145838,0.0004720494,0.0004289257,0.003215176,0.01988811,0.002054955],"genre_scores_gemma":[0.9091328,0.001422732,0.08443704,0.0001944815,0.00008525919,0.0001743431,0.003040439,0.000528527,0.0009842995],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009068538,"threshold_uncertainty_score":0.02376407,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03310799623204715,"score_gpt":0.3501939382944648,"score_spread":0.3170859420624177,"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."}}