{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.06020638,0.001616496,0.001170017,0.00763943,0.0008671047,0.005401761,0.002703478,0.001692392,0.0009202421],"category_scores_gemma":[0.2800744,0.0006307025,0.0008727176,0.003271863,0.0008592391,0.005216354,0.002610907,0.00149396,0.0009474859],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002544372,"about_ca_system_score_gemma":0.003567552,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006664138,"about_ca_topic_score_gemma":0.007066231,"domain_scores_codex":[0.9456758,0.03283033,0.005584608,0.004168293,0.01091743,0.0008235121],"domain_scores_gemma":[0.6268544,0.253558,0.02614821,0.02887514,0.06145912,0.00310518],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001174518,0.0009023637,0.1566326,0.0009462424,0.0004178572,0.0001693886,0.003195452,0.09219743,0.008149317,0.002959576,0.009857561,0.7233976],"study_design_scores_gemma":[0.0002790095,0.002105557,0.07541633,0.0002592886,0.0003269765,0.0004813404,0.001735033,0.8548175,0.0419348,0.00624359,0.01610857,0.0002918601],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7501807,0.003284073,0.2204907,0.002036571,0.0002942208,0.001002599,0.002539817,0.01347929,0.006691996],"genre_scores_gemma":[0.7838063,0.000409805,0.2089282,0.0001647446,0.00007963338,0.000268559,0.004852165,0.00048631,0.001004249],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06020638,"threshold_uncertainty_score":0.3184056,"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."}}