Background: Staphylococcus (S) aureus colonization is known to cause skin barrier disruption in atopic dermatitis (AD) patients. However, it has not been studied how S. aureus induces aberrant epidermal lipid composition and skin barrier dysfunction.Methods: Skin tape strips (STS) and swabs were obtained from 24 children with AD (6.0 ± 4.4 years) and 16 healthy children (7.0 ± 4.5 years). Lipidomic analysis of STS samples was performed by mass spectrometry. Skin levels of methicillin‐sensitive and methicillin‐resistant S. aureus (MSSA and MRSA) were evaluated. The effects of MSSA and MRSA were evaluated in primary human keratinocytes (HEKs) and organotypic skin cultures.Results: AD and organotypic skin colonized with MRSA significantly increased the proportion of lipid species with nonhydroxy fatty acid sphingosine ceramide with palmitic acid ([N‐16:0 NS‐CER], sphingomyelins [16:0–18:0 SM]), and lysophosphatidylcholines [16:0–18:0 LPC], but significantly reduced the proportion of corresponding very long‐chain fatty acids (VLCFAs) species (C22–28) compared to the skin without S. aureus colonization. Significantly increased transepidermal water loss (TEWL) was found in MRSA‐colonized AD skin. S. aureus indirectly through interleukin (IL)‐1β, tumor necrosis factor (TNF)‐α, IL‐6, and IL‐33 inhibited expression of fatty acid elongase enzymes (ELOVL3 and ELOVL4) in HEKs. ELOVL inhibition was more pronounced by MRSA and resulted in TEWL increase in organotypic skin.Conclusion: Aberrant skin lipid profiles and barrier dysfunction are associated with S. aureus colonization in AD patients. These effects are attributed to the inhibition of ELOVLs by S. aureus‐induced IL‐1β, TNF‐α, IL‐6, and IL‐33 seen in keratinocyte models and are more prominent in MRSA than MSSA.

Background: The diagnostic performance of convolutional neural networks (CNNs) for diagnosing several types of skin neoplasms has been demonstrated as comparable with that of dermatologists using clinical photography. However, the generalizability should be demonstrated using a large-scale external dataset that includes most types of skin neoplasms. In this study, the performance of a neural network algorithm was compared with that of dermatologists in both real-world practice and experimental settings. Methods and findings: To demonstrate generalizability, the skin cancer detection algorithm (https://rcnn.modelderm.com) developed in our previous study was used without modification. We conducted a retrospective study with all single lesion biopsied cases (43 disorders; 40,331 clinical images from 10,426 cases: 1,222 malignant cases and 9,204 benign cases); mean age (standard deviation [SD], 52.1 [18.3]; 4,701 men [45.1%]) were obtained from the Department of Dermatology, Severance Hospital in Seoul, Korea between January 1, 2008 and March 31, 2019. Using the external validation dataset, the predictions of the algorithm were compared with the clinical diagnoses of 65 attending physicians who had recorded the clinical diagnoses with thorough examinations in real-world practice. In addition, the results obtained by the algorithm for the data of randomly selected batches of 30 patients were compared with those obtained by 44 dermatologists in experimental settings; the dermatologists were only provided with multiple images of each lesion, without clinical information. With regard to the determination of malignancy, the area under the curve (AUC) achieved by the algorithm was 0.863 (95% confidence interval [CI] 0.852–0.875), when unprocessed clinical photographs were used. The sensitivity and specificity of the algorithm at the predefined high-specificity threshold were 62.7% (95% CI 59.9–65.1) and 90.0% (95% CI 89.4–90.6), respectively. Furthermore, the sensitivity and specificity of the first clinical impression of 65 attending physicians were 70.2% and 95.6%, respectively, which were superior to those of the algorithm (McNemar test; p < 0.0001). The positive and negative predictive values of the algorithm were 45.4% (CI 43.7–47.3) and 94.8% (CI 94.4–95.2), respectively, whereas those of the first clinical impression were 68.1% and 96.0%, respectively. In the reader test conducted using images corresponding to batches of 30 patients, the sensitivity and specificity of the algorithm at the predefined threshold were 66.9% (95% CI 57.7–76.0) and 87.4% (95% CI 82.5–92.2), respectively. Furthermore, the sensitivity and specificity derived from the first impression of 44 of the participants were 65.8% (95% CI 55.7–75.9) and 85.7% (95% CI 82.4–88.9), respectively, which are values comparable with those of the algorithm (Wilcoxon signed-rank test; p = 0.607 and 0.097). Limitations of this study include the exclusive use of high-quality clinical photographs taken in hospitals and the lack of ethnic diversity in the study population. Conclusions: Our algorithm could diagnose skin tumors with nearly the same accuracy as a dermatologist when the diagnosis was performed solely with photographs. However, as a result of limited data relevancy, the performance was inferior to that of actual medical examination. To achieve more accurate predictive diagnoses, clinical information should be integrated with imaging information.

Kim J, et al. Allergy Asthma Immunol Res. 2023 Mar;15(2):186-200. https://doi.org/10.4168/aair.2023.15.2.186

Recently, the number of nationwide medical researches on psoriasis using the National Health Insurance Service database has been on the rise. However, identification of psoriasis using diagnostic codes alone can lead to misclassification. Accuracy of the diagnostic codes and their concordance with medical records should be validated first to identify psoriasis patients correctly. To validate the diagnostic codes of psoriasis (International Classification of Diseases, 10th Revision L40) and to find the algorithm for the identification of psoriasis. We collected medical records of patients who received their first diagnostic codes of psoriasis during 5 years from five hospitals. Fifteen percent of psoriasis patients were randomly selected from each hospital. We performed a validation by reviewing medical records and compared 5 algorithms to identify the best algorithm. Total of 538 cases were reviewed and classified as psoriasis (n=368), not psoriasis (n=159), and questionable (n=11). The most accurate algorithm was including patients with ≥1 visits with psoriasis as primary diagnostic codes and prescription of vitamin D derivatives. Its positive predictive value was 96.5% (95% confidence interval [CI], 93.9%~98.1%), which was significantly higher than those of the algorithm, including patients with ≥1 visits with psoriasis as primary diagnostic codes or including ≥1 visits with diagnostic codes of psoriasis (primary or additional) (91.0% and 69.8%). Sensitivity was 90.8% (95% CI, 87.2%~93.4%) and specificity was 92.5% (95% CI, 86.9%~95.9%). Our study demonstrates a validated algorithm to identify psoriasis, which will be useful for the nationwide population-based study of psoriasis in Korea.