{"id":"W2963824827","doi":"10.2196/14083","title":"Developing a Standardization Algorithm for Categorical Laboratory Tests for Clinical Big Data Research: Retrospective Study","year":2019,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Electronic Health Records Systems","field":"Health Professions","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Standardization; Categorical variable; Computer science; Data mining; Algorithm; Data science; Medicine; Medical physics; Machine learning","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","research_integrity"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.03128054,0.0002371615,0.0008766811,0.0002512772,0.0008839393,0.00003774006,0.001091148,0.0008040805,0.00006938286],"category_scores_gemma":[0.01067399,0.0001920655,0.00006815836,0.0009372773,0.0001364929,0.0004225941,0.0006582898,0.002320233,0.0002756593],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001671416,"about_ca_system_score_gemma":0.01420325,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006085455,"about_ca_topic_score_gemma":0.0003679245,"domain_scores_codex":[0.9914395,0.001500915,0.003088573,0.0004501742,0.002056069,0.001464796],"domain_scores_gemma":[0.9885126,0.006340221,0.0007070894,0.001290001,0.002501865,0.0006481882],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0004957464,0.001181765,0.1815538,0.006713728,0.0003927056,0.000008545998,0.02991669,7.826598e-7,0.000001230562,0.009101729,0.4161329,0.3545004],"study_design_scores_gemma":[0.009333496,0.003229563,0.009058543,0.000704972,0.00003485373,0.00000353691,0.03291541,0.07972997,0.000001537211,0.001245941,0.8632953,0.0004469211],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07349876,0.0002824309,0.8709018,0.003306926,0.008949199,0.03985704,0.001489793,0.0004289614,0.001285052],"genre_scores_gemma":[0.7019895,0.00180652,0.2072653,0.01755907,0.0278836,0.03191279,0.005895269,0.0007218239,0.004966164],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6636365,"threshold_uncertainty_score":0.9999815,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5084334879671147,"score_gpt":0.6169982422761427,"score_spread":0.108564754309028,"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."}}