{"id":"W4411477178","doi":"10.1101/2025.06.18.25329494","title":"Artificial Intelligence for Pre-Anaemic Iron Deficiency Detection Using Rich Complete Blood Count Data","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Erythropoietin and Anemia Treatment","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Nutrasource","funders":"NIHR Cambridge Biomedical Research Centre; Economic and Social Research Council; University of Cambridge; NIHR BioResource; Chief Scientist Office, Scottish Government Health and Social Care Directorate; Medical Research Council; Public Health Agency; Department of Health and Social Care; Scottish Government; British Heart Foundation; Engineering and Physical Sciences Research Council; Health and Social Care Research and Development Division; National Institute for Health and Care Research; NHS Blood and Transplant","keywords":"Complete blood count; Analyser; Iron deficiency; Count data; Blood count; Medicine; Anemia; Pediatrics; Internal medicine; Statistics; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.003039657,0.0005337394,0.0004776795,0.001972638,0.0002440468,0.001650466,0.0004258791,0.0005615138,0.002008387],"category_scores_gemma":[0.01284133,0.0001818566,0.0004601416,0.001047779,0.0003857522,0.0006508562,0.0006973529,0.0009112544,0.0005986873],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004932273,"about_ca_system_score_gemma":0.0006248581,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002765083,"about_ca_topic_score_gemma":0.001779334,"domain_scores_codex":[0.9987385,0.000767133,0.00006710841,0.0001772011,0.000204678,0.0000454316],"domain_scores_gemma":[0.9926203,0.006285242,0.0003313098,0.0003225347,0.0003291974,0.0001113842],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001184225,0.0007200401,0.1427556,0.0008612686,0.0007737587,0.0006630082,0.0006056272,0.24997,0.01529651,0.01073111,0.01593218,0.5605067],"study_design_scores_gemma":[0.00003134118,0.0001552345,0.02277921,0.00007572737,0.00006072049,0.0001663771,0.0001433557,0.9554134,0.00304988,0.01483707,0.003259951,0.00002778229],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5918038,0.004447475,0.3704819,0.007717255,0.0003353088,0.0003606891,0.004835508,0.004587546,0.01543041],"genre_scores_gemma":[0.8858436,0.0006573452,0.1094196,0.0004242396,0.0001520798,0.0001070004,0.002051433,0.00004449461,0.001300282],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.003039657,"threshold_uncertainty_score":0.01607543,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1339609715083971,"score_gpt":0.3742564869745464,"score_spread":0.2402955154661493,"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."}}