{"id":"W4396975716","doi":"10.1186/s12911-024-02529-9","title":"BarlowTwins-CXR: enhancing chest X-ray abnormality localization in heterogeneous data with cross-domain self-supervised learning","year":2024,"lang":"en","type":"article","venue":"BMC Medical Informatics and Decision Making","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; Université de Montréal; Centre Intégré Universitaire de Santé et de Services Sociaux du Centre-Sud-de-l'Île-de-Montréal","funders":"","keywords":"Artificial intelligence; Computer science; Abnormality; Transfer of learning; Deep learning; Feature (linguistics); Domain (mathematical analysis); Pattern recognition (psychology); Intersection (aeronautics); Pyramid (geometry); Machine learning; Medicine","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":[0.002996822,0.001422595,0.0009792569,0.001438879,0.0004199831,0.0008160119,0.002163385,0.001274505,0.001296365],"category_scores_gemma":[0.004083743,0.0003783126,0.001094602,0.0008539184,0.000587251,0.001129212,0.001661751,0.001436003,0.0006811958],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006922052,"about_ca_system_score_gemma":0.001046837,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004693619,"about_ca_topic_score_gemma":0.00539312,"domain_scores_codex":[0.9985555,0.0003625746,0.00008023295,0.0005698298,0.0003109451,0.0001209136],"domain_scores_gemma":[0.9980683,0.0006221697,0.0002976772,0.0004283365,0.0004618529,0.0001216359],"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":[0.0008185918,0.001143009,0.01856061,0.0003954238,0.0005199023,0.0006268208,0.0002816985,0.2713823,0.02093579,0.001936995,0.02269627,0.6607025],"study_design_scores_gemma":[0.00003371997,0.0001602375,0.001686406,0.00001649631,0.0000246758,0.0001326412,0.0000251216,0.9903057,0.005514167,0.0009960866,0.001091374,0.00001345335],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1937592,0.001644018,0.7815164,0.0005154772,0.0002103564,0.0003885668,0.001102587,0.01842001,0.002443287],"genre_scores_gemma":[0.7460132,0.0002829685,0.2449804,0.0005367855,0.0001366367,0.000240818,0.004138438,0.0005035533,0.003167226],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004693619,"threshold_uncertainty_score":0.01584888,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03718349212735889,"score_gpt":0.3610629034154232,"score_spread":0.3238794112880644,"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."}}