DePicT Melanoma Deep-CLASS: a deep convolutional neural networks approach to classify skin lesion images.
- 作者列表："Nasiri S","Helsper J","Jung M","Fathi M
BACKGROUND:Melanoma results in the vast majority of skin cancer deaths during the last decades, even though this disease accounts for only one percent of all skin cancers' instances. The survival rates of melanoma from early to terminal stages is more than fifty percent. Therefore, having the right information at the right time by early detection with monitoring skin lesions to find potential problems is essential to surviving this type of cancer. RESULTS:An approach to classify skin lesions using deep learning for early detection of melanoma in a case-based reasoning (CBR) system is proposed. This approach has been employed for retrieving new input images from the case base of the proposed system DePicT Melanoma Deep-CLASS to support users with more accurate recommendations relevant to their requested problem (e.g., image of affected area). The efficiency of our system has been verified by utilizing the ISIC Archive dataset in analysis of skin lesion classification as a benign and malignant melanoma. The kernel of DePicT Melanoma Deep-CLASS is built upon a convolutional neural network (CNN) composed of sixteen layers (excluding input and ouput layers), which can be recursively trained and learned. Our approach depicts an improved performance and accuracy in testing on the ISIC Archive dataset. CONCLUSIONS:Our methodology derived from a deep CNN, generates case representations for our case base to use in the retrieval process. Integration of this approach to DePicT Melanoma CLASS, significantly improving the efficiency of its image classification and the quality of the recommendation part of the system. The proposed method has been tested and validated on 1796 dermoscopy images. Analyzed results indicate that it is efficient on malignancy detection.
背景: 在过去的几十年里，黑色素瘤导致了绝大多数的皮肤癌死亡，尽管这种疾病只占所有皮肤癌病例的 1%。黑色素瘤从早期到晚期的生存率超过 50%。因此，在正确的时间通过监测皮肤病变的早期检测以发现潜在的问题来获得正确的信息对于这种类型的癌症的生存至关重要。 结果: 提出了一种在基于案例推理 (CBR) 系统中使用深度学习对皮肤病变进行早期检测的方法。这种方法已经被用于从所提出的系统的base中检索新的输入图像，描述黑色素瘤深类，以支持用户获得与其请求的问题相关的更准确的建议 (e。g.，疫区的图像)。通过利用ISIC存档数据集分析良恶性黑色素瘤的皮损分类，验证了我们系统的效率。描绘黑色素瘤深度类的内核是建立在由 16 层 (不包括输入和输出层) 组成的卷积神经网络 (CNN) 上的，可以递归地训练和学习。我们的方法描述了在ISIC存档数据集上测试的性能和准确性的改进。 结论: 我们的方法来源于深度CNN，为我们的病例base生成病例表示，用于检索过程。整合这种方法来描述黑色素瘤类别，显著提高了其图像分类的效率和系统推荐部分的质量。该方法已在 1796 幅皮肤镜图像上进行了测试和验证。分析结果表明，它对恶性肿瘤的检测是有效的。
METHODS::Blue rubber bleb naevus syndrome (BRBNS) is an extremely rare venous malformation that often manifests as multiple haemangioma-like lesions in the skin and gastrointestinal tract. The drug sirolimus plays a key role in the signalling pathway of angiogenesis and subsequent development of BRBNS and its use has been described in several case reports. We present a case series of four patients with BRBNS who exhibited good treatment response to sirolimus. All four patients were administered oral sirolimus at doses of 1.0-1.5 mg/m2 /day with a target drug level of 5-10 ng/mL and median treatment duration of 20 months. All patients had a reduction in the size of the lesions and a normalization of coagulopathy with tolerable drug adverse reactions at follow-up. Sirolimus may be effective and safe in paediatric patients with BRBNS. Further prospective studies are suggested to evaluate the long-term effectiveness of this drug.
METHODS:BACKGROUND:Human papillomavirus (HPV) infections are associated with common dermatologic and nondermatologic diseases. Although HPV vaccines are well established as preventive measures for genital warts and cervical neoplasia, their use as therapeutic agents deserves greater attention. OBJECTIVE:To evaluate the use of HPV vaccine(s) as a treatment modality for cutaneous and/or mucosal disease. METHODS:A primary literature search using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines was conducted in January 2019 by using the PubMed and Cochrane databases. RESULTS:A total of 63 articles with 4439 patients were included. The majority of patients with cutaneous warts, recurrent respiratory papillomatosis, and squamous and basal cell carcinomas were successfully treated with HPV vaccination. Preliminary data on patients with pre-existing anogenital warts, cervical intraepithelial neoplasia, anal intraepithelial neoplasia, and vulvar intraepithelial neoplasia is promising. LIMITATIONS:This review was limited by the lack of controls, patients' previous HPV vaccination status, and publication bias. CONCLUSION:The commercially available three-dose, quadrivalent HPV vaccine is a potential therapeutic option for the treatment of cutaneous warts, recurrent respiratory papillomatosis, and squamous and basal cell carcinomas. Noncommercially available HPV vaccines demonstrate therapeutic response for treating anogenital warts, cervical intraepithelial neoplasia, anal intraepithelial neoplasia, and vulvar intraepithelial neoplasia. The vaccine's efficacy as an adjunct therapy for HPV-associated cutaneous and/or mucosal disease warrants further exploration.
METHODS::Our understanding of melanoma precursors and progression to melanoma has developed as a result of advances in the field of molecular diagnostics. We now better understand the potential for genetic heterogeneity within a single lesion. Combined tumors can pose a diagnostic challenge when deciding the line between benign and malignant, which in turn has direct implications for patient management. Primary cilia (PC) are ubiquitous sensory organelles that have essential functions in cellular proliferation, differentiation, and development. The ciliation index (percentage of ciliated melanocytes) has been shown to reliably differentiate melanoma, which fail to ciliate, from melanocytic nevi, which retain PC. We therefore analyzed the potential for using the ciliation index to differentiate benign and malignant components in combined melanocytic lesions. We collected patient samples (n = 10) of unequivocal combined lesions with both melanoma and associated nevus components. Melanocytes were highlighted with SOX10 and costained with gamma-Tubulin and acetylated alpha-Tubulin to highlight the basal body and cilium, respectively. The number of melanocytes retaining cilia under high-power microscopy was examined. The melanoma component had average of 4% ciliation (SD: 7%), whereas the associated nevus component was significantly higher with 59% ciliation (SD: 17%). These data show that PC may be a reliable means of distinguishing benign from malignant components within a single tumor. The ciliation index may be a helpful tool in distinguishing challenging cases of combined lesions of melanoma in situ with a dermal nevus component from invasive melanoma, thus promoting improved staging and clinical management.