{"id":"W2770017713","doi":"10.1186/s41512-018-0028-3","title":"Using ordinal outcomes to construct and select biomarker combinations for single-level prediction","year":2018,"lang":"en","type":"article","venue":"Diagnostic and Prognostic Research","topic":"Sepsis Diagnosis and Treatment","field":"Medicine","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Abbott Diagnostics; Pritzker School of Medicine; University of California, San Francisco; National Institutes of Health; National Institute of Diabetes and Digestive and Kidney Diseases; National Heart, Lung, and Blood Institute; Institute for Clinical Evaluative Sciences; U.S. Department of Veterans Affairs","keywords":"Outcome (game theory); Logistic regression; Biomarker; Context (archaeology); Computer science; Selection (genetic algorithm); Construct (python library); Machine learning; Artificial intelligence; Data mining; Statistics; 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":[0.02460515,0.002478328,0.002252101,0.006939532,0.0008663602,0.002839214,0.001408365,0.0009988553,0.002983298],"category_scores_gemma":[0.06357998,0.0006979093,0.002250741,0.004215037,0.0009589453,0.00217771,0.00268423,0.002768688,0.0007908488],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009678698,"about_ca_system_score_gemma":0.002031573,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001119597,"about_ca_topic_score_gemma":0.00172076,"domain_scores_codex":[0.9879727,0.00775131,0.001134014,0.001303141,0.001543293,0.0002954147],"domain_scores_gemma":[0.9413252,0.04382671,0.00678677,0.003366348,0.003735676,0.0009594137],"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":[0.002100159,0.000883544,0.2679758,0.0008581048,0.002328597,0.0004435694,0.0006642037,0.2284837,0.005290275,0.00830238,0.004514882,0.4781549],"study_design_scores_gemma":[0.0003556771,0.0009940958,0.02849405,0.0002826932,0.0007106914,0.0004065493,0.000255413,0.9008104,0.005865056,0.05790861,0.003714649,0.00020201],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08894332,0.0006382087,0.9063482,0.000644081,0.00008354711,0.0006697503,0.0007463857,0.0007791686,0.001147327],"genre_scores_gemma":[0.3964166,0.0002561286,0.6006636,0.0001420312,0.0001007706,0.0008163894,0.001134486,0.0001029229,0.0003670615],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02460515,"threshold_uncertainty_score":0.1301261,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5615408223281894,"score_gpt":0.5013232235527673,"score_spread":0.0602175987754221,"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."}}