{"id":"W4377115686","doi":"10.1038/s41598-023-35230-4","title":"Accurate drusen segmentation in optical coherence tomography via order-constrained regression of retinal layer heights","year":2023,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Medizinische Fakultät, Rheinische Friedrich-Wilhelms-Universität Bonn; Rheinische Friedrich-Wilhelms-Universität Bonn","keywords":"Drusen; Optical coherence tomography; Segmentation; Artificial intelligence; Computer science; Ground truth; Macular degeneration; Retinal; Multispectral image; Pattern recognition (psychology); Ophthalmology; Medicine","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":[0.0005229877,0.0007592023,0.0004547024,0.0008562433,0.0002158175,0.0007705934,0.0006418313,0.0006442567,0.0009225084],"category_scores_gemma":[0.002027514,0.0003669287,0.0005534168,0.0005447008,0.0002950952,0.0006747864,0.0006348535,0.0007106934,0.0005978884],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006425802,"about_ca_system_score_gemma":0.0009716345,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01115304,"about_ca_topic_score_gemma":0.01908977,"domain_scores_codex":[0.9997202,0.00005496255,0.00001420676,0.00009523977,0.00007694503,0.00003848211],"domain_scores_gemma":[0.9995195,0.000155281,0.0001041628,0.00007472672,0.0001145226,0.00003192962],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004742402,0.0002238841,0.01932587,0.0002644004,0.0002506639,0.000310786,0.0001883826,0.413307,0.09702501,0.003188808,0.009280416,0.4561605],"study_design_scores_gemma":[0.000007980505,0.00001759406,0.002389078,0.00001105526,0.00001431774,0.00005988793,0.00001014238,0.9888903,0.006991981,0.0009971128,0.0005984652,0.00001204692],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2360142,0.001618708,0.7530589,0.0005698114,0.00007024864,0.00008985246,0.0009911692,0.005086435,0.002500671],"genre_scores_gemma":[0.7558621,0.001012871,0.2356251,0.0004016296,0.00006649599,0.0001007907,0.0019719,0.0005064955,0.004452574],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01115304,"threshold_uncertainty_score":0.02217627,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02478618940444063,"score_gpt":0.318546266255758,"score_spread":0.2937600768513174,"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."}}