{"id":"W4399821708","doi":"10.3389/fonc.2024.1320220","title":"Improving skin cancer detection by Raman spectroscopy using convolutional neural networks and data augmentation","year":2024,"lang":"en","type":"article","venue":"Frontiers in Oncology","topic":"Spectroscopy Techniques in Biomedical and Chemical Research","field":"Biochemistry, Genetics and Molecular Biology","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"BC Children's Hospital; Vancouver Coastal Health Research Institute; University of British Columbia; Vancouver Coastal Health","funders":"Canadian Institutes of Health Research; BC Hydro; Canadian Dermatology Foundation; Canadian Cancer Society; VGH and UBC Hospital Foundation; University of British Columbia","keywords":"Receiver operating characteristic; Artificial intelligence; Linear discriminant analysis; Pattern recognition (psychology); Convolutional neural network; Principal component analysis; Support vector machine; Computer science; Raman spectroscopy; Pooling; Basal cell carcinoma; Mathematics; Machine learning; Medicine; Pathology; Physics; Basal cell; Optics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007897425,0.0009010187,0.0003920018,0.0006441453,0.000176503,0.0004071415,0.000575872,0.0004448502,0.0007444744],"category_scores_gemma":[0.001519392,0.000263556,0.0007969665,0.0004039253,0.0003346405,0.0005647289,0.0005339617,0.0005482231,0.000336083],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004885813,"about_ca_system_score_gemma":0.0003871756,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00208631,"about_ca_topic_score_gemma":0.003409872,"domain_scores_codex":[0.999726,0.00006324728,0.00001422298,0.0000832614,0.00007710693,0.00003615115],"domain_scores_gemma":[0.9993659,0.0002962689,0.00009919943,0.00007610826,0.0001426601,0.00001997828],"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":[0.0005633506,0.0004322604,0.01862253,0.0003574732,0.0002327687,0.0002905159,0.000105278,0.2903293,0.2012725,0.001133655,0.001900239,0.4847602],"study_design_scores_gemma":[0.000009581073,0.0001471663,0.003847258,0.00001687434,0.00006095437,0.0001382738,0.00001389023,0.9417951,0.0521973,0.0006996069,0.001054729,0.00001933288],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4696045,0.002169821,0.5219927,0.0004835328,0.000105533,0.0001078144,0.0004026369,0.002585831,0.002547701],"genre_scores_gemma":[0.8405771,0.000714868,0.156642,0.0001437736,0.00003747477,0.00009167376,0.000452372,0.00007127749,0.001269383],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00208631,"threshold_uncertainty_score":0.004176617,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01601570372987302,"score_gpt":0.3742931622299826,"score_spread":0.3582774585001096,"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."}}