{"id":"W4398224430","doi":"10.1016/j.jacr.2024.04.027","title":"Establishing a Validation Infrastructure for Imaging-Based Artificial Intelligence Algorithms Before Clinical Implementation","year":2024,"lang":"en","type":"review","venue":"Journal of the American College of Radiology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":23,"is_retracted":false,"has_abstract":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Cancer Institute; National Institutes of Health; Amazon Web Services; University of Washington; Institute for Data Intensive Research in Astrophysics and Cosmology, University of Washington; American Cancer Society","keywords":"Workflow; Generalizability theory; Computer science; Artificial intelligence; Machine learning; Algorithm; Database","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":[0.06434284,0.001702682,0.003302364,0.004696239,0.0008227191,0.007635634,0.006952073,0.003995433,0.004888856],"category_scores_gemma":[0.113201,0.001309962,0.002501257,0.002478517,0.003746388,0.00806486,0.005033393,0.007300687,0.003995898],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003564224,"about_ca_system_score_gemma":0.02034643,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005650551,"about_ca_topic_score_gemma":0.003895401,"domain_scores_codex":[0.9742861,0.01021209,0.003583969,0.002048205,0.009180881,0.0006887775],"domain_scores_gemma":[0.8740287,0.06864509,0.006469215,0.01289478,0.03652793,0.00143428],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001494117,0.0001761076,0.002015402,0.008648383,0.0003181579,0.00009439063,0.0001763419,0.003113927,0.001471627,0.04639398,0.03168691,0.9057553],"study_design_scores_gemma":[0.000217766,0.0006318606,0.008460129,0.04237769,0.0009434405,0.0009670872,0.0003092317,0.02397814,0.0108409,0.07638348,0.834658,0.0002322752],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"review","genre_scores_codex":[0.005501049,0.4525034,0.4581016,0.03486857,0.005538231,0.002985578,0.001679165,0.004200199,0.03462226],"genre_scores_gemma":[0.07259872,0.2846642,0.605491,0.01503726,0.004008116,0.003175068,0.008665155,0.000989847,0.005370605],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.9356571,"threshold_uncertainty_score":0.3402815,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1659381653970396,"score_gpt":0.5241187030360461,"score_spread":0.3581805376390065,"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."}}