{"id":"W4388926194","doi":"10.48550/arxiv.2311.11647","title":"Cancer-Net PCa-Data: An Open-Source Benchmark Dataset for Prostate Cancer Clinical Decision Support using Synthetic Correlated Diffusion Imaging Data","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"MRI in cancer diagnosis","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Prostate cancer; Artificial intelligence; Cancer; Computer science; Magnetic resonance imaging; Principal component analysis; Benchmark (surveying); Pattern recognition (psychology); Medicine; Radiology; Internal medicine","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["open_science"],"consensus_categories":[],"category_scores_codex":[0.002236821,0.001984152,0.001075548,0.002640628,0.0007845276,0.001976354,0.003746981,0.002700791,0.00842778],"category_scores_gemma":[0.01301342,0.0006123389,0.001718,0.003477416,0.0007542292,0.0008589587,0.001723875,0.002071414,0.007246723],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002043474,"about_ca_system_score_gemma":0.003481854,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02714286,"about_ca_topic_score_gemma":0.04201956,"domain_scores_codex":[0.9981915,0.0004535293,0.0001907433,0.0004809617,0.000516366,0.0001669171],"domain_scores_gemma":[0.99661,0.001500301,0.0002283049,0.0005902665,0.0007550149,0.0003160858],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001200124,0.0007148682,0.01264765,0.002117811,0.0006158548,0.0006161621,0.0001352265,0.05219268,0.002620461,0.003680495,0.8346145,0.08884428],"study_design_scores_gemma":[0.00226922,0.0007915216,0.03423996,0.0006801877,0.0003978976,0.003339614,0.0003924332,0.3550905,0.0166173,0.02473609,0.5610753,0.0003701128],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.05630795,0.006043752,0.03895781,0.00531023,0.0009105089,0.001088843,0.858583,0.02339773,0.009400228],"genre_scores_gemma":[0.06103709,0.001065305,0.03871839,0.0008053069,0.0001695787,0.0007086052,0.8941126,0.001000327,0.002382761],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.996253,"threshold_uncertainty_score":0.05396974,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3199061833152542,"score_gpt":0.377216053527165,"score_spread":0.05730987021191086,"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."}}