{"id":"W4214835785","doi":"10.1101/2022.02.28.482322","title":"Cytologic Scoring of Equine Exercise-Induced Pulmonary Hemorrhage (EIPH): Performance of Human Experts and a Deep Learning-Based Algorithm","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Clinical Laboratory Practices and Quality Control","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Hemosiderin; Ground truth; Reproducibility; Bronchoalveolar lavage; Artificial intelligence; Algorithm; Computer science; Grading (engineering); Medicine; Machine learning; Pathology; Statistics; Internal medicine; Mathematics; Lung","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001907782,0.0006589267,0.001878149,0.0004178934,0.0002673655,0.00005166949,0.0004409904,0.0005814295,0.0002977139],"category_scores_gemma":[0.0007314164,0.0006745726,0.0003062301,0.0006249329,0.000249181,0.0001879488,0.0007509722,0.001706716,0.000003838411],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001874339,"about_ca_system_score_gemma":0.0005897863,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008892096,"about_ca_topic_score_gemma":0.000001333232,"domain_scores_codex":[0.9955117,0.0004117467,0.001524732,0.001172803,0.0007989153,0.0005801002],"domain_scores_gemma":[0.9957262,0.0004021588,0.001330048,0.001392473,0.0007612824,0.0003877997],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0009462673,0.001326131,0.06062533,0.005425104,0.000275142,0.0004767344,0.00003929053,0.0002110237,0.9301988,0.00004125651,0.000006301223,0.0004286605],"study_design_scores_gemma":[0.01004333,0.00868304,0.2081053,0.005465186,0.003117894,9.447344e-7,0.0001831151,0.1531312,0.6065676,0.000003661071,0.001627878,0.003070841],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9926064,0.005006293,0.0002841172,0.0001678074,0.0004668247,0.001149915,0.00009037343,0.00021109,0.00001720128],"genre_scores_gemma":[0.9944438,0.0006279679,0.004101826,0.0001801615,0.0002467627,0.0002532828,0.000002918976,0.000134481,0.00000875047],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3236311,"threshold_uncertainty_score":0.9995705,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03646978360720748,"score_gpt":0.297253587778162,"score_spread":0.2607838041709545,"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."}}