{"id":"W4391871139","doi":"10.1016/j.jwpe.2024.105692","title":"Evaluation of activated sludge settling characteristics from microscopy images with deep convolutional neural networks and transfer learning","year":2024,"lang":"en","type":"preprint","venue":"Journal of Water Process Engineering","topic":"AI in cancer detection","field":"Computer Science","cited_by":40,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université Laval","funders":"Vlaamse regering; Fundação de Amparo à Pesquisa do Estado de São Paulo","keywords":"Convolutional neural network; Transfer of learning; Settling; Deep learning; Artificial intelligence; Computer science; Activated sludge; Microscopy; Artificial neural network; Materials science; Environmental science; Physics; Optics; Sewage treatment; Environmental engineering","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":[],"consensus_categories":[],"category_scores_codex":[0.0008918308,0.0002474568,0.0003812435,0.0002415773,0.00004452341,0.0002211973,0.0003143374,0.0001604874,0.000005260929],"category_scores_gemma":[0.00003265657,0.0001871974,0.00006950704,0.0001270373,0.00002936845,0.0003711012,0.0001759898,0.00131897,2.449545e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001895262,"about_ca_system_score_gemma":0.0001462212,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007702419,"about_ca_topic_score_gemma":7.123718e-7,"domain_scores_codex":[0.9982136,0.00006695538,0.000525979,0.0002894607,0.0006955426,0.0002084834],"domain_scores_gemma":[0.9986678,0.00005372324,0.0002169756,0.0001316407,0.000863076,0.00006672424],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003917335,0.000009150287,0.0001631079,0.0003507839,0.0001835457,0.00001000145,0.001251193,0.8994816,0.09392849,0.000002431517,0.00000113415,0.004579423],"study_design_scores_gemma":[0.0003280401,0.0000863824,0.0006654091,0.0007851342,0.0002398807,0.00008178448,0.00002103554,0.8492525,0.1480988,0.000251532,0.000005175741,0.0001843108],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.562265,0.0009794735,0.4359056,0.00006401744,0.0006577952,0.00008774205,0.000003932124,0.00003437059,0.000002035669],"genre_scores_gemma":[0.9942204,0.00007044889,0.00526196,0.000005983147,0.0003817744,0.00001132039,0.00001078333,0.00003504466,0.000002336533],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4319553,"threshold_uncertainty_score":0.7633688,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01144453071109233,"score_gpt":0.2443246662430125,"score_spread":0.2328801355319201,"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."}}