{"id":"W2890268332","doi":"10.1007/978-3-030-00934-2_12","title":"A Multitask Learning Architecture for Simultaneous Segmentation of Bright and Red Lesions in Fundus Images","year":2018,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":38,"is_retracted":false,"has_abstract":false,"ca_institutions":"Hôpital Maisonneuve-Rosemont; Polytechnique Montréal","funders":"","keywords":"Computer science; Segmentation; Pooling; Artificial intelligence; Decoding methods; Encoding (memory); Pattern recognition (psychology); Image segmentation; Encoder; Task (project management); Residual; Algorithm","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":[0.001116708,0.001144417,0.001064544,0.0007149755,0.0005849584,0.0009331501,0.002269143,0.002181924,0.003868933],"category_scores_gemma":[0.001020711,0.0006797956,0.001185429,0.000901576,0.000340893,0.0009433846,0.001529956,0.001590693,0.001921055],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008234963,"about_ca_system_score_gemma":0.001141428,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01381097,"about_ca_topic_score_gemma":0.01856411,"domain_scores_codex":[0.9997119,0.00004505486,0.00001625067,0.0001144087,0.00004952633,0.00006279814],"domain_scores_gemma":[0.9994848,0.0001755199,0.00002856452,0.00007270732,0.000181746,0.00005662258],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0005544083,0.0003832408,0.001514643,0.0001487577,0.0002390626,0.0001393302,0.0000905505,0.1064794,0.04044843,0.001968266,0.01118049,0.8368534],"study_design_scores_gemma":[0.00001502366,0.00006748598,0.0006648198,0.00001097284,0.00004487153,0.00004385131,0.00001151289,0.9891232,0.006831767,0.002126634,0.001047525,0.00001234735],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05080362,0.001423741,0.9375048,0.0005198392,0.0002695183,0.0001065955,0.0006141382,0.006194947,0.002562822],"genre_scores_gemma":[0.47262,0.0009389805,0.5058892,0.0007321923,0.0003339335,0.0002560236,0.002310511,0.0004103771,0.01650886],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01381097,"threshold_uncertainty_score":0.02746117,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01497406942873735,"score_gpt":0.2872900682464993,"score_spread":0.272315998817762,"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."}}