{"id":"W4362731331","doi":"10.1186/s41747-023-00330-3","title":"Evaluation of techniques to improve a deep learning algorithm for the automatic detection of intracranial haemorrhage on CT head imaging","year":2023,"lang":"en","type":"article","venue":"European Radiology Experimental","topic":"Intracerebral and Subarachnoid Hemorrhage Research","field":"Medicine","cited_by":29,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Melbourne","keywords":"Convolutional neural network; Neuroradiology; Receiver operating characteristic; Artificial intelligence; Preprocessor; Deep learning; Computer science; Algorithm; Pattern recognition (psychology); Radiology; Medicine; Machine learning; Neurology","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.003467218,0.001399628,0.0006081116,0.001123116,0.0002527119,0.0009475739,0.001505537,0.001000441,0.001246954],"category_scores_gemma":[0.008008076,0.00039786,0.0007388052,0.0006397545,0.0002711792,0.001024926,0.0009571093,0.0008563951,0.0006272323],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001235603,"about_ca_system_score_gemma":0.001312072,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008491468,"about_ca_topic_score_gemma":0.009021895,"domain_scores_codex":[0.9990151,0.000247389,0.000118104,0.0002746431,0.0002407886,0.0001039243],"domain_scores_gemma":[0.9966272,0.001445824,0.0002633973,0.0002970205,0.001272043,0.00009448338],"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":[0.0007843547,0.0005114811,0.01694055,0.0003240987,0.0003651537,0.0001477961,0.00007240204,0.2521534,0.03195676,0.0007389502,0.003276744,0.6927283],"study_design_scores_gemma":[0.00004424217,0.0002884712,0.002192394,0.00002288847,0.00007029995,0.00006907086,0.00001575532,0.9746529,0.02154861,0.0002609201,0.0008235248,0.00001078844],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4772207,0.002777032,0.5053539,0.0007633618,0.00024015,0.0005030062,0.0007539418,0.009514467,0.002873464],"genre_scores_gemma":[0.7012966,0.0004872516,0.2944694,0.0003670565,0.00004627314,0.0001787833,0.00125414,0.000191705,0.001708919],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008491468,"threshold_uncertainty_score":0.01833659,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02953317234107794,"score_gpt":0.3497645436354063,"score_spread":0.3202313712943284,"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."}}