Systematic outperformance of 112 dermatologists in multiclass skin cancer image classification by convolutional neural networks.

BACKGROUND: Recently, convolutional neural networks (CNNs) systematically outperformed dermatologists in distinguishing dermoscopic melanoma and nevi images. However, such a binary classification does not reflect the clinical reality of skin cancer screenings in which multiple diagnoses need to be taken into account. METHODS: Using 11,444 dermoscopic images, which covered dermatologic diagnoses comprising the majority of commonly pigmented skin lesions commonly faced in skin cancer screenings, a CNN was trained through novel deep learning techniques. A test set of 300 biopsy-verified images was used to compare the classifier's performance with that of 112 dermatologists from 13 German university hospitals. The primary end-point was the correct classification of the different lesions into benign and malignant. The secondary end-point was the correct classification of the images into one of the five diagnostic categories. FINDINGS: Sensitivity and specificity of dermatologists for the primary end-point were 74.4% (95% confidence interval [CI]: 67.0-81.8%) and 59.8% (95% CI: 49.8-69.8%), respectively. At equal sensitivity, the algorithm achieved a specificity of 91.3% (95% CI: 85.5-97.1%). For the secondary end-point, the mean sensitivity and specificity of the dermatologists were at 56.5% (95% CI: 42.8-70.2%) and 89.2% (95% CI: 85.0-93.3%), respectively. At equal sensitivity, the algorithm achieved a specificity of 98.8%. Two-sided McNemar tests revealed significance for the primary end-point (p < 0.001). For the secondary end-point, outperformance (p < 0.001) was achieved except for basal cell carcinoma (on-par performance). INTERPRETATION: Our findings show that automated classification of dermoscopic melanoma and nevi images is extendable to a multiclass classification problem, thus better reflecting clinical differential diagnoses, while still outperforming dermatologists at a significant level (p < 0.001).

A 2q24.2 microdeletion containing TANK as novel candidate gene for intellectual disability.

Interstitial deletions within the chromosomal region 2q24.2 have already been linked to intellectual disability (ID) in the past. In most cases the described patients showed a syndromic form of ID associated with large deletions containing multiple genes. Here we describe a family with two siblings with mild non-syndromic ID. They shared the same 564 kb deletion in the chromosomal region 2q24.2 containing only the TANK gene, which was inherited from the similarly affected father, thus suggesting haploinsufficiency of TANK as a novel cause of non-syndromic ID. TANK encodes the TRAF family member-associated NF-kappa-B activator (OMIM #603893), which is expressed in many tissues. It functions as an adapter protein that interacts with the NF-kappa-B pathway and SOX11, an essential transcription factor in regeneration, survival and differentiation of the neuronal system. TANK has not been linked to ID or other human diseases before. To further elucidate the role of TANK in non-syndromic ID, we screened a cohort of 288 TANK deletion negative non-syndromic mental retardation patients for TANK mutations without identifying any pathogenic variant.

Acquired intestinal aganglionosis after a lytic infection with varicella-zoster virus.

BACKGROUND AND PURPOSE: In this report, we present the first case of an immunologically impaired child surviving a lytic varicella-zoster virus infection affecting the enteric nervous system. In histological findings, myenteric and submucous enteric ganglia were nearly completely absent owing to virus infection. METHODS: A 3-year-old girl with acute lymphoblastic leukemia and generalized varicella-zoster infection developed an ileus. She underwent multiple laparotomies in which histological sections of the entire small intestine could be obtained. RESULTS: The histological evaluation of these samples showed a generalized aganglionosis with inflammatory residuals. A more detailed immunohistochemical analysis using neuronal (PGP, enolase), glial (S100), and lymphocytic (LCA) antibodies demonstrated a nearly complete neuronal loss. CONCLUSION: To our knowledge, this is the first case of a secondary intestinal aganglionosis after varicella-zoster virus infection.