Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists.

Background: Deep learning convolutional neural networks (CNN) may facilitate melanoma detection, but data comparing a CNN's diagnostic performance to larger groups of dermatologists are lacking. Methods: Google's Inception v4 CNN architecture was trained and validated using dermoscopic images and corresponding diagnoses. In a comparative cross-sectional reader study a 100-image test-set was used (level-I: dermoscopy only; level-II: dermoscopy plus clinical information and images). Main outcome measures were sensitivity, specificity and area under the curve (AUC) of receiver operating characteristics (ROC) for diagnostic classification (dichotomous) of lesions by the CNN versus an international group of 58 dermatologists during level-I or -II of the reader study. Secondary end points included the dermatologists' diagnostic performance in their management decisions and differences in the diagnostic performance of dermatologists during level-I and -II of the reader study. Additionally, the CNN's performance was compared with the top-five algorithms of the 2016 International Symposium on Biomedical Imaging (ISBI) challenge. Results: In level-I dermatologists achieved a mean (±standard deviation) sensitivity and specificity for lesion classification of 86.6% (±9.3%) and 71.3% (±11.2%), respectively. More clinical information (level-II) improved the sensitivity to 88.9% (±9.6%, P = 0.19) and specificity to 75.7% (±11.7%, P < 0.05). The CNN ROC curve revealed a higher specificity of 82.5% when compared with dermatologists in level-I (71.3%, P < 0.01) and level-II (75.7%, P < 0.01) at their sensitivities of 86.6% and 88.9%, respectively. The CNN ROC AUC was greater than the mean ROC area of dermatologists (0.86 versus 0.79, P < 0.01). The CNN scored results close to the top three algorithms of the ISBI 2016 challenge. Conclusions: For the first time we compared a CNN's diagnostic performance with a large international group of 58 dermatologists, including 30 experts. Most dermatologists were outperformed by the CNN. Irrespective of any physicians' experience, they may benefit from assistance by a CNN's image classification. Clinical trial number: This study was registered at the German Clinical Trial Register (DRKS-Study-ID: DRKS00013570; https://www.drks.de/drks_web/).

Furuncular Myiasis of the Lower Leg.

A 35-year-old Caucasian woman, otherwise healthy, presented with a four weeks history of painful, inflammatory nodules, each with a central opening on her right lower leg (Figure 1). Intermittently, a marked serosanguinous secretion was noted. The remaining skin and mucosa were not affected. The patient denied any history of trauma. One month earlier she had returned from a journey to Peru. Despite topical treatment with corticosteroids and antibiotics as well as systemic therapy with oral doxycycline, the lesions did not show any regression or reduction in their secretion. An incisional biopsy was performed at all sites and the extracted organic specimens were submitted for histopathological assessment. Histopathology revealed a mixed inflammatory dermal infiltrate consisting of numerous eosinophils admixed with some histiocytes and lymphocytes. Furthermore, an organic foreign body with an eosinophilic cuticle consistent with DH became evident (Figure 2, a-e). Human myiasis represents the infestation of humans by developing dipterous larvae (maggots) of various fly species (1). The most common flies causing human myiasis are Dermatobia hominis (DH, "human botfly") and, less frequently, Cordylobia anthropophaga ("tumbu fly") or Cordylobia rodhaini (1). DH is indigenous in Central and South America, but in Europe and the United States of America only cases of travelers that imported the infestation have been described (2,3). The adult DH is a yellow-headed fly with a grey-blue body of approximately 15 mm. DH is active throughout the year and predominantly found in humid and high temperature regions of Central and South America (2). The larvae of DH develop as obligate parasites in living tissue, causing furuncular myiasis (2,4). Clinically crusted nodules with serosanguinous secretion and sometimes nocturnal pain affecting uncovered anatomical sites are typical (2,4). Differential diagnoses include bacterial furunculosis, arthropod bite reaction, pyoderma, inflamed epidermoid cysts, or cutaneous leishmaniasis (2,5-7). Most cases of furuncular myiasis can be diagnosed solely by their clinical presentation and after carefully taking the patients' history. Nevertheless, dermoscopy and sonography may help to confirm the diagnosis and rule out the abovementioned differential diagnoses (6,7). Although myiasis is a self-limited disease, it can be accompanied by severe complications including tetanus or bacterial superinfections (2). The literature reports a number of treatment options including a) occlusion of the skin opening by paraffin oil, beeswax, petroleum jelly or agents such as liquid nitrogen or ethyl chloride sprays to suffocate the larva and promote self-extrusion; b) removing the larva by forceps after lidocaine instillation; c) surgical removal or d) the use of topical or oral ivermectin (1,8,9). Preventive measures include protective clothing, insect repellents, and sleeping curtains. Histopathology of furuncular myiasis typically reveals a mixed inflammatory infiltrate of eosinophils, lymphocytes, and macrophages. Sometimes formation of granulomas with giant cells becomes evident (10). Fly larvae show a thick eosinophilic cuticle with external facing sclerotized spines. Internal structures (respiratory and digestive tracts or striated muscle) may occasionally be identified (10), (Figure 2, b-e). In most cases, tissue sections will not be sufficient to identify the fly's genus or its species; however, identification by stereomicroscopy has been reported (10). In conclusion, we described the clinical and histopathological features of furuncular myiasis by DH in a traveler returning from Peru. Given the ever-increasing numbers of international travelers, western dermatopathologists and dermatologists should be familiar with this disease to avoid prescription of unnecessary topical or systemic medications.

Cutaneous leiomyosarcoma with myxoid alteration arising in a setting of multiple cutaneous smooth muscle neoplasms.

BACKGROUND: Cutaneous leiomyomas and leiomyosarcomas are rare tumors that originate from the arrector muscle of hair follicles or the smooth muscle of blood vessels. CASE REPORT: A 74-year-old male presented with a single, erythematous nodule on the left upper arm. This lesion developed within the excision scar of a piloleiomyoma that had been excised 3 years ago. Additionally, physical examination revealed a tender nodule on the right ear. RESULTS: Histological examination showed a cutaneous leiomyosarcoma with myxoid alteration on the upper arm and an angioleiomyoma on the right ear. CONCLUSION: Myxoid leiomyosarcomas have exclusively been reported in the uterus and gastrointestinal tract, but not in the skin. Here, we describe a case of cutaneous leiomyosarcoma with myxoid alteration that developed in the excision scar of a piloleiomyoma and was accompanied by a cutaneous angioleiomyoma in a different location.

Novel stabilin-1 interacting chitinase-like protein (SI-CLP) is up-regulated in alternatively activated macrophages and secreted via lysosomal pathway.

Mammalian Glyco_18-domain-containing proteins include catalytically active chitinases and chitinase-like proteins with cytokine activity involved in host defense and Th2-type inflammatory reactions. Here, we describe a novel human Glyco_18-domain-containing protein, SI-CLP, as an interacting partner of the endocytic/sorting receptor stabilin-1. Similarly to the chitinase-like cytokines YKL-39, YKL-40, and YM1/2, SI-CLP lacks a chitin-binding domain and catalytic amino acids. Using a novel mAb 1C11, we demonstrated that SI-CLP is sorted into late endosomes and secretory lysosomes in human alternatively activated macrophages. The direct interaction of SI-CLP with stabilin-1, their colocalization in the trans-Golgi network, and the reduced sorting of SI-CLP into lysosomes in macrophages treated with stabilin-1 siRNA suggest that stabilin-1 is involved in intracellular sorting of SI-CLP. Expression of SI-CLP in macrophages was strongly up-regulated by the Th2 cytokine IL-4 and by dexamethasone. This effect was suppressed by IFNgamma but not affected by IL-10. In contrast, expression of YKL-40 was induced by IFNgamma and suppressed by dexamethasone. Macrophages treated with IL-4 secreted SI-CLP, while costimulation with dexamethasone blocked secretion and resulted in intracellular accumulation of SI-CLP. The 1C11 mAb detected SI-CLP in human bronchoalveolar lavage and peripheral-blood leukocytes (PBLs), and can be used to analyze the role of SI-CLP in human disorders.