{"id":"W1994343730","doi":"10.1002/sim.1723","title":"Three validation metrics for automated probabilistic image segmentation of brain tumours","year":2004,"lang":"en","type":"article","venue":"Statistics in Medicine","topic":"Medical Image Segmentation Techniques","field":"Computer Science","cited_by":95,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Center for Research Resources; U.S. National Library of Medicine; National Cancer Institute; Agency for Healthcare Research and Quality; National Institutes of Health","keywords":"Segmentation; Computer science; Artificial intelligence; Markov random field; Metric (unit); Similarity (geometry); Pixel; Pattern recognition (psychology); Gold standard (test); Image segmentation; Image (mathematics); Mathematics; Statistics","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":[],"consensus_categories":[],"category_scores_codex":[0.001226667,0.0001293075,0.0002650124,0.000406982,0.00003938193,0.00002284226,0.0003824552,0.00005007071,0.00002569916],"category_scores_gemma":[0.005076148,0.0001157317,0.00001836797,0.0008944515,0.0002170033,0.0002392813,0.00005241206,0.0001040133,0.000003156427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001902891,"about_ca_system_score_gemma":0.00013326,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001378961,"about_ca_topic_score_gemma":0.00004582826,"domain_scores_codex":[0.9982594,0.00004889233,0.0006402569,0.0002837477,0.0005632891,0.0002044422],"domain_scores_gemma":[0.9980875,0.000911828,0.0002898672,0.0002845069,0.0003448807,0.00008141777],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00008487922,0.0008168269,0.00135401,0.00178819,0.00006842419,0.000102675,0.006617204,0.001850796,0.1630574,0.5376045,0.07506582,0.2115894],"study_design_scores_gemma":[0.005846266,0.001493818,0.006550688,0.0004421081,0.00005873552,0.00001311132,0.000249772,0.1980303,0.1366986,0.6502268,0.00002962844,0.0003601898],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0008811906,0.00002998245,0.9960353,0.001542444,0.0002201821,0.000893524,0.00005757825,0.0002595743,0.00008027285],"genre_scores_gemma":[0.02993931,0.00001375037,0.9693884,0.0003179647,0.00003720583,0.00009663254,0.0001831219,0.00001241673,0.00001115267],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2112292,"threshold_uncertainty_score":0.607699,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02776990431574732,"score_gpt":0.3596305153886217,"score_spread":0.3318606110728743,"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."}}