{"id":"W4229071840","doi":"10.1016/j.jcct.2022.04.005","title":"Head to head comparison reproducibility and inter-reader agreement of an AI based coronary stenosis algorithm vs level 3 readers","year":2022,"lang":"en","type":"letter","venue":"Journal of cardiovascular computed tomography","topic":"Cardiac Imaging and Diagnostics","field":"Medicine","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"St. Paul's Hospital","funders":"","keywords":"Medicine; Reproducibility; Head (geology); Stenosis; Cardiology; Radiology; Internal medicine; Statistics; Mathematics; Geology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.03168257,0.000291885,0.001000027,0.0008938391,0.001129206,0.002399758,0.001139657,0.003821137,0.004499112],"category_scores_gemma":[0.1488198,0.0003959714,0.001039512,0.0006974646,0.001004837,0.001362087,0.000946796,0.002010784,0.002793926],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001532079,"about_ca_system_score_gemma":0.000915053,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003162351,"about_ca_topic_score_gemma":0.005954199,"domain_scores_codex":[0.9746674,0.0124154,0.003578,0.003179086,0.005208557,0.0009516007],"domain_scores_gemma":[0.7834069,0.1421073,0.005554197,0.0133616,0.05424546,0.001324471],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.03003677,0.0005962726,0.507479,0.0007498245,0.002381261,0.005701555,0.00927639,0.001717266,0.02394842,0.003032723,0.1302345,0.284846],"study_design_scores_gemma":[0.001088395,0.005200362,0.8211414,0.0005731473,0.002329713,0.01648884,0.005866303,0.01751665,0.03972431,0.009028153,0.08052958,0.0005131274],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.818075,0.004436853,0.03337338,0.07220478,0.02010038,0.0006081649,0.002579888,0.001038522,0.04758299],"genre_scores_gemma":[0.9690806,0.00030482,0.007804873,0.01196829,0.002436849,0.0001953227,0.0004816554,0.0002465929,0.007480968],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9683174,"threshold_uncertainty_score":0.1675555,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04254665842602113,"score_gpt":0.3019439822592798,"score_spread":0.2593973238332586,"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."}}