{"id":"W4385546024","doi":"10.1016/j.patter.2023.100802","title":"Evaluating progress in automatic chest X-ray radiology report generation","year":2023,"lang":"en","type":"article","venue":"Patterns","topic":"Topic Modeling","field":"Computer Science","cited_by":151,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Ontario; University of Toronto","funders":"National Institute of Biomedical Imaging and Bioengineering; National Institutes of Health; International Business Machines Corporation; National Heart, Lung, and Blood Institute; Gordon and Betty Moore Foundation","keywords":"Workload; Computer science; Metric (unit); Correctness; Selection (genetic algorithm); Artificial intelligence; Medical physics; Machine learning; Radiology; Data mining; Medicine; Algorithm; Engineering","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009716657,0.00007308213,0.0001128831,0.0001210429,0.00004258094,0.00005857785,0.0003063823,0.00004234611,0.0000140851],"category_scores_gemma":[0.00007999529,0.00007060427,0.00002220554,0.0002492063,0.000009545635,0.000155162,0.0001387954,0.00008198142,0.0000584911],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003686145,"about_ca_system_score_gemma":0.00004465855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002564491,"about_ca_topic_score_gemma":0.00003047997,"domain_scores_codex":[0.998845,0.0001055229,0.0002968617,0.0003345859,0.0001940803,0.0002239878],"domain_scores_gemma":[0.9993747,0.00004418306,0.00008755556,0.0004368264,0.00002643148,0.00003033735],"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":[3.560641e-7,0.00002589315,0.6152083,0.0000569423,0.000008692788,0.0005446928,0.002728412,0.01101053,0.001219366,0.0006011996,0.0001661033,0.3684295],"study_design_scores_gemma":[0.0001018688,0.00001610559,0.1812724,0.00001829546,0.000001198576,0.00004863059,0.00001264827,0.8181634,0.0001272926,0.0001548184,0.0000184481,0.00006487707],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7546906,0.00002375994,0.2436835,0.0008198137,0.0004035686,0.0001243158,2.792995e-7,0.0002147069,0.00003946468],"genre_scores_gemma":[0.970315,0.000002026915,0.0292599,0.00008554656,0.0001663992,0.00007959265,0.00001238489,0.000006421094,0.00007270193],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8071529,"threshold_uncertainty_score":0.2879159,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1259875085762207,"score_gpt":0.3731220470944587,"score_spread":0.2471345385182381,"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."}}