{"id":"W4406160197","doi":"10.1186/s13040-024-00412-x","title":"The Venus score for the assessment of the quality and trustworthiness of biomedical datasets","year":2025,"lang":"en","type":"article","venue":"BioData Mining","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"HORIZON EUROPE European Research Council; Ministero dell'Università e della Ricerca; HORIZON EUROPE Framework Programme; Dipartimenti di Eccellenza; European Commission; Alexander von Humboldt-Stiftung","keywords":"Computer science; Quality (philosophy); Informatics; Usability; Data science; Data quality; Data mining; Artificial intelligence","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":[0.02848014,0.001287313,0.001599048,0.02318514,0.002062644,0.004844851,0.001644428,0.002081063,0.005876746],"category_scores_gemma":[0.2102576,0.000353339,0.002226879,0.01611891,0.002680213,0.005166089,0.005121651,0.00178713,0.001872723],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001917384,"about_ca_system_score_gemma":0.002599001,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002504474,"about_ca_topic_score_gemma":0.004197825,"domain_scores_codex":[0.9474834,0.01993802,0.009596187,0.004265033,0.01730839,0.001409026],"domain_scores_gemma":[0.7859559,0.151469,0.01669301,0.01222613,0.02853164,0.005124252],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.002194742,0.0003810025,0.4865975,0.006190182,0.002899563,0.0006445413,0.004277576,0.01419181,0.005135243,0.0374819,0.07453058,0.3654754],"study_design_scores_gemma":[0.0008650593,0.002713442,0.4484774,0.003635588,0.001867705,0.003393948,0.007798791,0.1352389,0.0177666,0.184291,0.1922736,0.001677967],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5358184,0.0106261,0.3203368,0.004180203,0.001846484,0.00387387,0.05481544,0.006384134,0.06211862],"genre_scores_gemma":[0.8075184,0.001124446,0.1591123,0.0006344563,0.0002591086,0.003054981,0.02509042,0.0009070197,0.002298862],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9715198,"threshold_uncertainty_score":0.1506192,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3350027862965513,"score_gpt":0.5494247029763654,"score_spread":0.214421916679814,"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."}}