{"id":"W6981072173","doi":"","title":"Development of a Composite Health Index in Children with Cystic Fibrosis: A Pipeline for Data Processing, Machine Learning, and Model Implementation using Electronic Health Records","year":2022,"lang":"en","type":"dissertation","venue":"UCL Discovery (University College London)","topic":"Cystic Fibrosis Research Advances","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Interpretability; Cluster (spacecraft); Respiratory Medicine; Pipeline (software); Feature (linguistics); Cluster analysis; Field (mathematics); Confusion matrix","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0006035278,0.000339869,0.0008600454,0.0007978352,0.000609329,0.00003069083,0.0003074143,0.00008006904,0.00003014772],"category_scores_gemma":[0.00003402148,0.000370143,0.00005470058,0.0007861899,0.00005813201,0.0006562081,0.0002367001,0.0005989748,1.765484e-7],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001733143,"about_ca_system_score_gemma":0.008694,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003344011,"about_ca_topic_score_gemma":0.03892228,"domain_scores_codex":[0.9972503,0.0001460755,0.0005297663,0.0007927274,0.0006471096,0.0006340747],"domain_scores_gemma":[0.9985108,0.00005970991,0.0008185763,0.0003078495,0.0001275288,0.0001755468],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.1612763,0.005648215,0.4004977,0.04836389,0.003622434,0.0001008151,0.04007559,0.02071002,0.007142722,0.001522336,0.0035665,0.3074735],"study_design_scores_gemma":[0.01837512,0.002997422,0.06632827,0.002890643,0.0004077627,0.0001015956,0.02330849,0.8803427,0.0001005368,0.00006068311,0.004168301,0.0009184993],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9457874,0.004536771,0.04100554,0.0002371244,0.00002794461,0.003829902,0.004515216,0.00004154679,0.00001849757],"genre_scores_gemma":[0.858633,0.001009548,0.07296246,0.00007887858,0.00002829902,0.0000362439,0.06225093,0.0001183291,0.004882277],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8596327,"threshold_uncertainty_score":0.9998751,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01771626155112457,"score_gpt":0.3277759636090533,"score_spread":0.3100597020579287,"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."}}