{"id":"W2150196286","doi":"10.1093/bib/bbr008","title":"Caveats and pitfalls of ROC analysis in clinical microarray research (and how to avoid them)","year":2011,"lang":"en","type":"article","venue":"Briefings in Bioinformatics","topic":"AI in cancer detection","field":"Computer Science","cited_by":134,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Receiver operating characteristic; Ranking (information retrieval); Metric (unit); Gold standard (test); Confidence interval; Statistics; Performance metric; Computer science; Machine learning; Mathematics; 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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.3953603,0.002319342,0.003197479,0.005117199,0.003658378,0.009335922,0.009167825,0.01097089,0.00258424],"category_scores_gemma":[0.6792365,0.001984339,0.003208547,0.007461883,0.03659594,0.01252626,0.006127669,0.02908234,0.001975218],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005257026,"about_ca_system_score_gemma":0.009266717,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008870915,"about_ca_topic_score_gemma":0.008498536,"domain_scores_codex":[0.6132197,0.2920917,0.03227321,0.01279458,0.04823195,0.001388894],"domain_scores_gemma":[0.2170925,0.7025673,0.02430372,0.02426966,0.03023331,0.001533477],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000707061,0.000166027,0.01171346,0.006775217,0.00139147,0.002198517,0.01330543,0.009196984,0.001332627,0.3964081,0.2365402,0.3202649],"study_design_scores_gemma":[0.0003130368,0.0002790492,0.005430639,0.01077628,0.000356433,0.003026983,0.003862627,0.01334198,0.001988274,0.8158193,0.144201,0.0006044217],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"commentary","genre_gemma":"methods","genre_scores_codex":[0.006120123,0.04782081,0.3631883,0.557733,0.01571567,0.0005332906,0.0005093948,0.001306342,0.007073036],"genre_scores_gemma":[0.1311931,0.02242746,0.4990213,0.3146126,0.02580151,0.002582193,0.0002095197,0.0008568437,0.003295554],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6046396,"threshold_uncertainty_score":0.7456284,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1267116161059898,"score_gpt":0.3407471400724225,"score_spread":0.2140355239664327,"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."}}