Saliency-Enhanced Content-Based Image Retrieval for Diagnosis Support in Dermatology Consultation: Reader Study.

BACKGROUND: Previous research studies have demonstrated that medical content image retrieval can play an important role by assisting dermatologists in skin lesion diagnosis. However, current state-of-the-art approaches have not been adopted in routine consultation, partly due to the lack of interpretability limiting trust by clinical users. OBJECTIVE: This study developed a new image retrieval architecture for polarized or dermoscopic imaging guided by interpretable saliency maps. This approach provides better feature extraction, leading to better quantitative retrieval performance as well as providing interpretability for an eventual real-world implementation. METHODS: Content-based image retrieval (CBIR) algorithms rely on the comparison of image features embedded by convolutional neural network (CNN) against a labeled data set. Saliency maps are computer vision-interpretable methods that highlight the most relevant regions for the prediction made by a neural network. By introducing a fine-tuning stage that includes saliency maps to guide feature extraction, the accuracy of image retrieval is optimized. We refer to this approach as saliency-enhanced CBIR (SE-CBIR). A reader study was designed at the University Hospital Zurich Dermatology Clinic to evaluate SE-CBIR's retrieval accuracy as well as the impact of the participant's confidence on the diagnosis. RESULTS: SE-CBIR improved the retrieval accuracy by 7% (77% vs 84%) when doing single-lesion retrieval against traditional CBIR. The reader study showed an overall increase in classification accuracy of 22% (62% vs 84%) when the participant is provided with SE-CBIR retrieved images. In addition, the overall confidence in the lesion's diagnosis increased by 24%. Finally, the use of SE-CBIR as a support tool helped the participants reduce the number of nonmelanoma lesions previously diagnosed as melanoma (overdiagnosis) by 53%. CONCLUSIONS: SE-CBIR presents better retrieval accuracy compared to traditional CBIR CNN-based approaches. Furthermore, we have shown how these support tools can help dermatologists and residents improve diagnosis accuracy and confidence. Additionally, by introducing interpretable methods, we should expect increased acceptance and use of these tools in routine consultation.

Clinical and Molecular Features of Skin Malignancies in Muir-Torre Syndrome.

BACKGROUND: We investigated the mutational landscape of skin tumors in patients with Muir-Torre Syndrome (MTS) a hereditary autosomal dominant mismatch repair disorder of increased cancer susceptibility, and examined mutations other than in the DNA mismatch repair (MMR) genes. METHODS: This retrospective single-center case series included seven patients with the diagnosis of Muir-Torre Syndrome with precise medical history and family history. Mutational analysis of tumor samples Formalin-fixed paraffin-embedded tissue blocks of skin lesions associated with Muir-Torre Syndrome were used for further analysis. All skin tumors were analyzed with the Oncomine Comprehensive Assay v3 (Life Technologies), which includes 161 of the most relevant cancer driver genes. RESULTS: Eleven skin neoplasms (nine sebaceous tumors, one melanoma, one cutaneous squamous cell carcinoma) were diagnosed in seven patients. In two patients, visceral malignancies preceded the diagnosis of the skin tumors and one patient was diagnosed with a visceral malignancy after a sebaceous tumor. History of familial cancer of Lynch Syndrome (LS) was reported in three patients. The most frequently detected mutation was in the MSH2 gene, followed by mutations in the NOTCH1/2 and TP53 gene. Conclusion, this study provides a molecular analysis of Muir-Torre Syndrome associated and non-associated skin tumors in patients with Muir-Torre Syndrome. Patients with sebaceous lesions should undergo microsatellite instability analysis and accurate evaluation of personal and family history to detect a possible Muir-Torre syndrome. As secondary malignancies may appear years after the first occurrence of sebaceous tumors, lifelong screening is mandatory.

Complications After Cosmetic Surgery Tourism.

Background: Cosmetic surgery tourism characterizes a phenomenon of people traveling abroad for aesthetic surgery treatment. Problems arise when patients return with complications or need of follow-up care. Objectives: To investigate the complications of cosmetic surgery tourism treated at our hospital as well as to analyze arising costs for the health system. Methods: Between 2010 and 2014, we retrospectively included all patients presenting with complications arising from cosmetic surgery abroad. We reviewed medical records for patients' characteristics including performed operations, complications, and treatment. Associated cost expenditure and Diagnose Related Groups (DRG)-related reimbursement were analyzed. Results: In total 109 patients were identified. All patients were female with a mean age of 38.5 ± 11.3 years. Most procedures were performed in South America (43%) and Southeast (29.4%) or central Europe (24.8%), respectively. Favored procedures were breast augmentation (39.4%), abdominoplasty (11%), and breast reduction (7.3%). Median time between the initial procedure abroad and presentation was 15 days (interquartile range [IQR], 9) for early, 81.5 days (IQR, 69.5) for midterm, and 4.9 years (IQR, 9.4) for late complications. Main complications were infections (25.7%), wound breakdown (19.3%), and pain/discomfort (14.7%). The majority of patients (63.3%) were treated conservatively; 34.8% became inpatients with a mean hospital stay of 5.2 ± 3.8 days. Overall DRG-related reimbursement premiums approximately covered the total costs. Conclusions: Despite warnings regarding associated risks, cosmetic surgery tourism has become increasingly popular. Efficient patients' referral to secondary/tertiary care centers with standardized evaluation and treatment can limit arising costs without imposing a too large burden on the social healthcare system. Level of Evidence: 4.