{"id":"W4411299320","doi":"10.1093/labmed/lmaf013","title":"Performance metrics for machine learning solutions in laboratory medicine","year":2025,"lang":"en","type":"review","venue":"Laboratory Medicine","topic":"Clinical Laboratory Practices and Quality Control","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Implementation; Computer science; Machine learning; Artificial intelligence; Software 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":["metaresearch","metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.008239449,0.001167006,0.006781223,0.002128506,0.0003707875,0.00002478815,0.0005632978,0.001180814,0.0005786016],"category_scores_gemma":[0.0336889,0.0008475403,0.0004238188,0.007840827,0.0006283702,0.0002920867,0.000173711,0.003238496,0.00006331313],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006461232,"about_ca_system_score_gemma":0.003510271,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007994306,"about_ca_topic_score_gemma":0.00009871305,"domain_scores_codex":[0.9917375,0.001334414,0.003457894,0.001352176,0.001011378,0.001106601],"domain_scores_gemma":[0.9850279,0.009791635,0.001565194,0.001321666,0.001641168,0.0006524057],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0006864304,0.0008169651,0.001552654,0.1577701,0.00124543,0.0002390738,0.0003299893,0.00001038305,0.00002350725,0.002954573,0.03299185,0.8013791],"study_design_scores_gemma":[0.005771176,0.001618966,0.00006789959,0.05260193,0.005294072,0.00001155551,0.0001458356,0.0005438311,8.391905e-7,0.00001633179,0.9333136,0.0006139711],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00002465714,0.9872081,0.0001454162,0.004438519,0.001747597,0.003413053,0.0006608801,0.0002508271,0.002110985],"genre_scores_gemma":[0.0001046099,0.989134,0.0003375927,0.003234118,0.002426699,0.0007249581,0.00124023,0.0001623675,0.002635456],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9003217,"threshold_uncertainty_score":0.9993975,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1386965396708843,"score_gpt":0.4423723086560829,"score_spread":0.3036757689851987,"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."}}