{"id":"W4392111888","doi":"10.1016/j.heliyon.2024.e26872","title":"Regional choroidal thickness estimation from color fundus images based on convolutional neural networks","year":2024,"lang":"en","type":"article","venue":"Heliyon","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Shantou Science and Technology Project; Natural Science Foundation of Guangdong Province; St. Thomas University; Science and Technology Planning Project of Guangdong Province; Shantou University","keywords":"Convolutional neural network; Mean absolute error; Fundus (uterus); Mean absolute percentage error; Artificial intelligence; Computer science; Correlation coefficient; Optical coherence tomography; Approximation error; Artificial neural network; Task (project management); Mathematics; Pattern recognition (psychology); Algorithm; Mean squared error; Ophthalmology; Statistics; Medicine","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001604408,0.0001525707,0.0002101423,0.0001379325,0.00009507529,0.00009147674,0.00005961791,0.00009430097,0.0003486502],"category_scores_gemma":[0.00007539455,0.000124496,0.0001573662,0.0002554755,0.00009372185,0.00009183937,0.0000133856,0.0003591014,0.0001449297],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009004193,"about_ca_system_score_gemma":0.0000855911,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006628362,"about_ca_topic_score_gemma":0.000003341695,"domain_scores_codex":[0.9988243,0.00006207191,0.0001999057,0.0003336988,0.0003801228,0.0001999003],"domain_scores_gemma":[0.9993048,0.0002722193,0.00003912082,0.0001845773,0.00009012113,0.0001091019],"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":[0.003818217,0.001131696,0.04820295,0.00199102,0.0009686556,0.002128837,0.0003961797,0.6802844,0.02677318,0.002600073,0.1078176,0.1238872],"study_design_scores_gemma":[0.0004670589,0.0001222339,0.02589229,0.0008103963,0.0001996187,0.00003802679,0.00002083323,0.9692815,0.0002186655,0.00006910512,0.002755624,0.0001246277],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7062877,0.02433579,0.1944178,0.06654342,0.001991381,0.0006164936,0.0001857589,0.001126225,0.004495507],"genre_scores_gemma":[0.9934632,0.0001050284,0.001614224,0.002103354,0.0007773643,0.00002011182,0.0005713225,0.00002436657,0.001320996],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2889971,"threshold_uncertainty_score":0.5076801,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02020958646951304,"score_gpt":0.2906705244946296,"score_spread":0.2704609380251166,"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."}}