{"id":"W3021508389","doi":"10.1109/jbhi.2020.2992878","title":"Multi-Task Pre-Training of Deep Neural Networks for Digital Pathology","year":2020,"lang":"en","type":"article","venue":"IEEE Journal of Biomedical and Health Informatics","topic":"AI in cancer detection","field":"Computer Science","cited_by":56,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Innovation for Defence Excellence and Security","keywords":"Feature extraction; Digital pathology; Feature (linguistics); Pattern recognition (psychology); Deep learning; Artificial neural network; Feature selection; Scheme (mathematics)","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":[0.001970576,0.001796042,0.0008925782,0.0007802295,0.0005211736,0.0009679612,0.002175842,0.001698056,0.005169452],"category_scores_gemma":[0.005275275,0.0007287574,0.001031129,0.0007900414,0.0006330675,0.002355359,0.002030326,0.003539184,0.002565313],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001013093,"about_ca_system_score_gemma":0.001481418,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004456706,"about_ca_topic_score_gemma":0.009046112,"domain_scores_codex":[0.9992969,0.0001426637,0.00003580351,0.0002117532,0.0001521167,0.0001606945],"domain_scores_gemma":[0.998163,0.0007771977,0.0001381344,0.0003978132,0.0004174943,0.0001063317],"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":[0.0008700844,0.001245903,0.005494474,0.0003837397,0.0002137262,0.0003718222,0.0002088394,0.3648579,0.03863359,0.003413002,0.01533459,0.5689724],"study_design_scores_gemma":[0.00003038434,0.0002742303,0.001545868,0.00003367923,0.00003506647,0.00009787712,0.00005804023,0.9665808,0.0241823,0.003779725,0.003358752,0.00002326427],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1623652,0.001339139,0.8178247,0.000794184,0.000349842,0.0003979551,0.000845615,0.01018945,0.005893933],"genre_scores_gemma":[0.7265272,0.0003735893,0.2571728,0.0009695956,0.0001342416,0.0005801825,0.003438013,0.0005108979,0.01029357],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005169452,"threshold_uncertainty_score":0.01729351,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05753932395557065,"score_gpt":0.3197855143351173,"score_spread":0.2622461903795467,"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."}}