{"id":"W4406407286","doi":"10.21203/rs.3.rs-5825843/v1","title":"IRMA: Machine learning-based harmonization of 18F-FDG PET brain scans in multi-center studies","year":2025,"lang":"en","type":"preprint","venue":"Research Square","topic":"Medical Imaging Techniques and Applications","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"National Institutes of Health; Stichting ParkinsonFonds; Canadian Institutes of Health Research; Ministero dell’Istruzione, dell’Università e della Ricerca; Foundation for the National Institutes of Health; Ministero della Salute; Alzheimer's Disease Neuroimaging Initiative; Alzheimer's Association; Northern California Institute for Research and Education","keywords":"Center (category theory); Harmonization; Pet imaging; Computer science; Artificial intelligence; Nuclear medicine; Medical physics; Medicine; Positron emission tomography; Physics; Chemistry","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":[],"consensus_categories":[],"category_scores_codex":[0.005688083,0.0009468352,0.001443071,0.003226263,0.0005796523,0.001875685,0.002077047,0.001002135,0.008318874],"category_scores_gemma":[0.01300885,0.0007331109,0.001522802,0.002989456,0.0003547675,0.001406345,0.00246176,0.0008177496,0.00293736],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004422469,"about_ca_system_score_gemma":0.0009745577,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001610997,"about_ca_topic_score_gemma":0.002815839,"domain_scores_codex":[0.9981383,0.000745477,0.0001157888,0.0005704039,0.0002873537,0.0001426189],"domain_scores_gemma":[0.9974865,0.001010464,0.0002932776,0.0007371571,0.0003560416,0.0001166947],"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":[0.001579577,0.0003268433,0.01163925,0.0003813915,0.001028996,0.0002444951,0.0002312338,0.03986582,0.01307578,0.003080592,0.03005524,0.8984907],"study_design_scores_gemma":[0.0005227289,0.000621862,0.03309704,0.00007669818,0.0006125252,0.001301844,0.0003544577,0.8831023,0.02544163,0.02399272,0.03073713,0.0001390398],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06712565,0.001511991,0.8917433,0.0005526074,0.0002942279,0.0006935563,0.005008006,0.0298119,0.003258705],"genre_scores_gemma":[0.259476,0.0004082227,0.7245914,0.0001958566,0.0002455739,0.0009860593,0.00713197,0.00386302,0.003101937],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008318874,"threshold_uncertainty_score":0.03008181,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1670262766802778,"score_gpt":0.5039607611824722,"score_spread":0.3369344845021944,"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."}}