{"id":"W3080569962","doi":"10.1073/pnas.2001227117","title":"Objective assessment of stored blood quality by deep learning","year":2020,"lang":"en","type":"article","venue":"Proceedings of the National Academy of Sciences","topic":"Erythrocyte Function and Pathophysiology","field":"Medicine","cited_by":104,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Blood Services; Toronto Metropolitan University; University of Alberta; Lunenfeld-Tanenbaum Research Institute; St. Michael's Hospital","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Research Councils UK; National Science Foundation; Government of Canada; Biotechnology and Biological Sciences Research Council; Canadian Blood Services","keywords":"Deep learning; Artificial intelligence; Computer science; Flow cytometry; Cytometry; Artificial neural network; Machine learning; Pathology; Pattern recognition (psychology); Biomedical engineering; Medicine; Immunology","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007319044,0.00006134619,0.0002130883,0.0000531222,0.00008568008,0.000003149997,0.000195683,0.00006363328,0.00003793652],"category_scores_gemma":[0.0009579117,0.00004071818,0.00007585054,0.0004846144,0.000510748,0.0001175189,0.00006431254,0.0002237955,5.040017e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001810882,"about_ca_system_score_gemma":0.00004830022,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000552624,"about_ca_topic_score_gemma":1.80723e-8,"domain_scores_codex":[0.9986592,0.00001513704,0.0002895496,0.0001792625,0.0007742434,0.00008262266],"domain_scores_gemma":[0.9990292,0.0001245013,0.0004682214,0.000004298391,0.0003315143,0.00004226333],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.00005043326,0.00007512969,0.04316416,0.0001268615,0.00003439479,3.024371e-9,0.0004681968,0.0000524155,0.9478173,0.007718326,0.0003169255,0.0001759154],"study_design_scores_gemma":[0.0008047908,0.0007045511,0.549051,0.00007393972,0.0000581523,0.00000506249,0.001688742,0.002289778,0.4422167,0.002698144,0.0003302978,0.00007893117],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9868794,0.00009657009,0.000008011084,0.004190455,0.00001346351,0.000132303,0.000007204178,0.00001092593,0.00866165],"genre_scores_gemma":[0.9977633,0.00003044217,0.001326443,0.0007255617,0.0000591434,0.000003738381,3.819127e-7,0.000002409715,0.00008862918],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5058868,"threshold_uncertainty_score":0.1881872,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06034704950259644,"score_gpt":0.3630150330555827,"score_spread":0.3026679835529862,"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."}}