{"id":"W4214536113","doi":"10.3389/fmats.2022.837006","title":"Training Deep Neural Networks to Reconstruct Nanoporous Structures From FIB Tomography Images Using Synthetic Training Data","year":2022,"lang":"en","type":"article","venue":"Frontiers in Materials","topic":"Electron and X-Ray Spectroscopy Techniques","field":"Materials Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Object Research Systems (Canada)","funders":"Deutsche Forschungsgemeinschaft","keywords":"Focused ion beam; Tomography; Convolutional neural network; Artificial intelligence; Materials science; Artificial neural network; Computer science; Segmentation; Voxel; Pattern recognition (psychology); Optics; Ion","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001521637,0.0003760478,0.0007826918,0.0003828221,0.0004406108,0.0003675131,0.00150998,0.0001094852,0.001837849],"category_scores_gemma":[0.0001088733,0.0003946688,0.00006472939,0.0004323698,0.0001662346,0.000400282,0.0006727399,0.0002768744,0.000001553887],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001825729,"about_ca_system_score_gemma":0.00007919539,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004735195,"about_ca_topic_score_gemma":0.00004349116,"domain_scores_codex":[0.9962168,0.0007372407,0.0007121242,0.00101097,0.0003843775,0.0009384894],"domain_scores_gemma":[0.9985216,0.00008606951,0.0002826133,0.0009508876,0.00002340232,0.0001354117],"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":[0.0002089226,0.00002075058,0.0005329852,0.00001223763,0.00002604844,0.00005390402,0.001873627,0.002424586,0.9794167,0.00007524357,0.001428963,0.01392599],"study_design_scores_gemma":[0.0008707331,0.0004516465,0.0008866061,0.0001175091,0.0001369139,0.0001776654,0.007413602,0.01858623,0.9495619,0.01850624,0.001670688,0.001620295],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9466428,0.0008654291,0.04526917,0.00009286608,0.005243992,0.0005559696,0.0009165773,0.0003153839,0.00009784118],"genre_scores_gemma":[0.7967953,0.00001606666,0.2022952,0.000306962,0.0002719923,0.00006754877,0.0001769045,0.00006193078,0.000008115721],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.157026,"threshold_uncertainty_score":0.9998505,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03313970600017286,"score_gpt":0.2828636451125926,"score_spread":0.2497239391124197,"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."}}