{"id":"W3121939634","doi":"10.1038/s41598-021-80998-y","title":"Improving prognostic performance in resectable pancreatic ductal adenocarcinoma using radiomics and deep learning features fusion in CT images","year":2021,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":60,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hospital for Sick Children; Lunenfeld-Tanenbaum Research Institute; Health Sciences Centre; Sinai Health System; Sunnybrook Health Science Centre; University of Alberta; University Health Network; University of Toronto","funders":"Canadian Cancer Society Research Institute; Terry Fox Research Institute; Pancreatic Cancer Canada Foundation; Hebrew University of Jerusalem; Government of Ontario; Ontario Institute for Cancer Research; Princess Margaret Cancer Foundation","keywords":"Radiomics; Feature (linguistics); Artificial intelligence; Deep learning; Transfer of learning; Medicine; Pancreatic ductal adenocarcinoma; Feature extraction; Computer science; Radiology; Machine learning; Pattern recognition (psychology); Internal medicine; Pancreatic cancer; Cancer","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":[],"consensus_categories":[],"category_scores_codex":[0.001868077,0.0001709594,0.0003742568,0.0003926143,0.0002746798,0.0002507118,0.00005799734,0.00005514737,0.00002740471],"category_scores_gemma":[0.002555996,0.0001577639,0.00004896191,0.0008041487,0.0002201512,0.0002414127,0.0001466987,0.0007540031,0.000001070222],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001515355,"about_ca_system_score_gemma":0.0003085171,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002765357,"about_ca_topic_score_gemma":0.00004331847,"domain_scores_codex":[0.9976918,0.0001122965,0.0005077096,0.0007838084,0.0004253244,0.0004791348],"domain_scores_gemma":[0.999099,0.00009259921,0.0001941262,0.0003574059,0.0000986935,0.0001582108],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00001935869,0.00004909964,0.8716756,0.0002683274,0.000007677584,0.009111655,0.0004672587,0.0008587981,0.1091709,0.000002902909,0.00002308761,0.008345386],"study_design_scores_gemma":[0.0007950112,0.00005873946,0.8044123,0.0009107194,0.00006689437,0.02044837,0.0006139455,0.1637769,0.008300266,0.0001203369,0.0002631807,0.0002333327],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9915354,0.006519853,0.0001275377,0.0001715701,0.0009751135,0.0002870732,1.755921e-7,0.00003662435,0.000346656],"genre_scores_gemma":[0.9938062,0.0001401796,0.004888533,0.00002994141,0.00008531667,0.00000899478,0.0000334222,0.00002828749,0.0009791472],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1629181,"threshold_uncertainty_score":0.6433424,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008462360894386671,"score_gpt":0.2542151189573353,"score_spread":0.2457527580629486,"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."}}