Search

Skin conditions affect an estimated 1.9 billion people worldwide. A shortage of dermatologists causes long wait times and leads patients to seek dermatologic care from general practitioners. However, the diagnostic accuracy of general practitioners has been reported to be only 0.24-0.70 (compared to 0.77-0.96 for dermatologists), resulting in referral errors, delays in care, and errors in diagnosis and treatment. In this paper, we developed a deep learning system (DLS) to provide a differential diagnosis of skin conditions for clinical cases (skin photographs and associated medical histories). The DLS distinguishes between 26 skin conditions that represent roughly 80% of the volume of skin conditions seen in primary care. The DLS was developed and validated using de-identified cases from a teledermatology practice serving 17 clinical sites via a temporal split: the first 14,021 cases for development and the last 3,756 cases for validation. On the validation set, where a panel of three board-certified dermatologists defined the reference standard for every case, the DLS achieved 0.71 and 0.93 top-1 and top-3 accuracies respectively. For a random subset of the validation set (n=963 cases), 18 clinicians reviewed the cases for comparison. On this subset, the DLS achieved a 0.67 top-1 accuracy, non-inferior to board-certified dermatologists (0.63, p

BACKGROUND Many dermatologic cases are first evaluated by primary care physicians or nurse practitioners. OBJECTIVE This study aimed to evaluate an artificial intelligence (AI)-based tool that assists with interpreting dermatologic conditions. METHODS We developed an AI-based tool and conducted a randomized multi-reader, multi-case study (20 primary care physicians, 20 nurse practitioners, and 1047 retrospective teledermatology cases) to evaluate its utility. Cases were enriched and comprised 120 skin conditions. Readers were recruited to optimize for geographical diversity; the primary care physicians practiced across 12 states (2-32 years of experience, mean 11.3 years), and the nurse practitioners practiced across 9 states (2-34 years of experience, mean 13.1 years). To avoid memory effects from incomplete washout, each case was read once by each clinician either with or without AI assistance, with the assignment randomized. The primary analyses evaluated the top-1 agreement, defined as the agreement rate of the clinicians’ primary diagnosis with the reference diagnoses provided by a panel of dermatologists (per case: 3 dermatologists from a pool of 12, practicing across 8 states, with 5-13 years of experience, mean 7.2 years of experience). We additionally conducted subgroup analyses stratified by cases’ self-reported race and ethnicity and measured the performance spread: the maximum performance subtracted by the minimum across subgroups. RESULTS The AI’s standalone top-1 agreement was 63%, and AI assistance was significantly associated with higher agreement with reference diagnoses. For primary care physicians, the increase in diagnostic agreement was 10% (PP CONCLUSIONS AI assistance was associated with significantly improved diagnostic agreement with dermatologists. Across race and ethnicity subgroups, for both primary care physicians and nurse practitioners, the effect of AI assistance remained high at 8%-15%, and the performance spread was similar at 5%-7%.

Background Many dermatologic cases are first evaluated by primary care physicians or nurse practitioners. Objective This study aimed to evaluate an artificial intelligence (AI)-based tool that assists with interpreting dermatologic conditions. Methods We developed an AI-based tool and conducted a randomized multi-reader, multi-case study (20 primary care physicians, 20 nurse practitioners, and 1047 retrospective teledermatology cases) to evaluate its utility. Cases were enriched and comprised 120 skin conditions. Readers were recruited to optimize for geographical diversity; the primary care physicians practiced across 12 states (2-32 years of experience, mean 11.3 years), and the nurse practitioners practiced across 9 states (2-34 years of experience, mean 13.1 years). To avoid memory effects from incomplete washout, each case was read once by each clinician either with or without AI assistance, with the assignment randomized. The primary analyses evaluated the top-1 agreement, defined as the agreement rate of the clinicians’ primary diagnosis with the reference diagnoses provided by a panel of dermatologists (per case: 3 dermatologists from a pool of 12, practicing across 8 states, with 5-13 years of experience, mean 7.2 years of experience). We additionally conducted subgroup analyses stratified by cases’ self-reported race and ethnicity and measured the performance spread: the maximum performance subtracted by the minimum across subgroups. Results The AI’s standalone top-1 agreement was 63%, and AI assistance was significantly associated with higher agreement with reference diagnoses. For primary care physicians, the increase in diagnostic agreement was 10% (P Conclusions AI assistance was associated with significantly improved diagnostic agreement with dermatologists. Across race and ethnicity subgroups, for both primary care physicians and nurse practitioners, the effect of AI assistance remained high at 8%-15%, and the performance spread was similar at 5%-7%. Acknowledgments This work was funded by Google LLC. Conflicts of Interest AJ, DW, VG, YG, GOM, JH, RS, CE, KN, KBD, GSC, LP, DRW, RCD, DC, Yun Liu, PB, and Yuan Liu are/were employees at Google and own Alphabet stocks.

Cardioviruses comprise a genus of picornaviruses that cause severe illnesses in rodents, but little is known about the prevalence, diversity, or spectrum of disease of such agents among humans. A single cardiovirus isolate, Saffold virus, was cultured in 1981 in stool from an infant with fever. Here, we describe the identification of a group of human cardioviruses that have been cloned directly from patient specimens, the first of which was detected using a pan-viral microarray in respiratory secretions from a child with influenza-like illness. Phylogenetic analysis of the nearly complete viral genome (7961 bp) revealed that this virus belongs to the Theiler's murine encephalomyelitis virus (TMEV) subgroup of cardioviruses and is most closely related to Saffold virus. Subsequent screening by RT-PCR of 719 additional respiratory specimens [637 (89%) from patients with acute respiratory illness] and 400 cerebrospinal fluid specimens from patients with neurological disease (aseptic meningitis, encephalitis, and multiple sclerosis) revealed no evidence of cardiovirus infection. However, screening of 751 stool specimens from 498 individuals in a gastroenteritis cohort resulted in the detection of 6 additional cardioviruses (1.2%). Although all 8 human cardioviruses (including Saffold virus) clustered together by phylogenetic analysis, significant sequence diversity was observed in the VP1 gene (66.9%–100% pairwise amino acid identities). These findings suggest that there exists a diverse group of novel human Theiler's murine encephalomyelitis virus-like cardioviruses that hitherto have gone largely undetected, are found primarily in the gastrointestinal tract, can be shed asymptomatically, and have potential links to enteric and extraintestinal disease.