Use of convolutional neural networks in skin lesion analysis using real world image and non-image data.

Background: Understanding performance of convolutional neural networks (CNNs) for binary (benign vs. malignant) lesion classification based on real world images is important for developing a meaningful clinical decision support (CDS) tool. Methods: We developed a CNN based on real world smartphone images with histopathological ground truth and tested the utility of structured electronic health record (EHR) data on model performance. Model accuracy was compared against three board-certified dermatologists for clinical validity. Results: At a classification threshold of 0.5, the sensitivity was 79 vs. 77 vs. 72%, and specificity was 64 vs. 65 vs. 57% for image-alone vs. combined image and clinical data vs. clinical data-alone models, respectively. The PPV was 68 vs. 69 vs. 62%, AUC was 0.79 vs. 0.79 vs. 0.69, and AP was 0.78 vs. 0.79 vs. 0.64 for image-alone vs. combined data vs. clinical data-alone models. Older age, male sex, and number of prior dermatology visits were important positive predictors for malignancy in the clinical data-alone model. Conclusion: Additional clinical data did not significantly improve CNN image model performance. Model accuracy for predicting malignant lesions was comparable to dermatologists (model: 71.31% vs. 3 dermatologists: 77.87, 69.88, and 71.93%), validating clinical utility. Prospective validation of the model in primary care setting will enhance understanding of the model's clinical utility.

Lesion identification and malignancy prediction from clinical dermatological images.

We consider machine-learning-based lesion identification and malignancy prediction from clinical dermatological images, which can be indistinctly acquired via smartphone or dermoscopy capture. Additionally, we do not assume that images contain single lesions, thus the framework supports both focal or wide-field images. Specifically, we propose a two-stage approach in which we first identify all lesions present in the image regardless of sub-type or likelihood of malignancy, then it estimates their likelihood of malignancy, and through aggregation, it also generates an image-level likelihood of malignancy that can be used for high-level screening processes. Further, we consider augmenting the proposed approach with clinical covariates (from electronic health records) and publicly available data (the ISIC dataset). Comprehensive experiments validated on an independent test dataset demonstrate that (1) the proposed approach outperforms alternative model architectures; (2) the model based on images outperforms a pure clinical model by a large margin, and the combination of images and clinical data does not significantly improves over the image-only model; and (3) the proposed framework offers comparable performance in terms of malignancy classification relative to three board certified dermatologists with different levels of experience.

Clinical and direct immunofluorescence characteristics of cutaneous toxicity associated with enfortumab vedotin.

Enfortumab vedotin (EV), a novel antibody-drug conjugate approved for metastatic urothelial carcinoma, causes a variety of cutaneous adverse reactions. We present two cases of bullous eruptions following treatment with EV, both demonstrating IgG deposition on direct immunofluorescence (DIF) correlating to the location of nectin-4 in the epidermis. This suggests that the IgG component of EV binding to nectin-4 in keratinocytes is likely a primary contributor to the high rates of cutaneous toxicity.

Cutaneous involvement by T-cell prolymphocytic leukemia presenting as livedoid vasculopathy.

T-cell prolymphocytic leukemia (T-PLL) is a rare, aggressive neoplasm derived from post-thymic T-cells. Patients are typically middle aged with a slight male predominance who present with a high white blood cell count, hepatosplenomegaly, lymphadenopathy, and other symptoms typically associated with leukemia. Although cutaneous involvement has been reported in up to 30% of cases of T-PLL, to our knowledge, none have presented with a presentation resembling livedoid vasculopathy. In the correct clinical context, an underlying hematolymphoid neoplasm should be included in the differential diagnosis of a patient presenting with livedoid vasculopathy.

Role of imaging in low-grade cutaneous B-cell lymphoma presenting in the skin.

BACKGROUND: Whole-body imaging is the current standard of care for staging all patients presenting with skin lesions of B-cell lymphomas (BCLs), regardless of skin disease extent; however, supporting data are lacking. OBJECTIVE: To determine the clinical utility of imaging in the detection of systemic involvement in low-grade cutaneous BCLs in the skin. METHODS: Retrospective cohort analysis of patients presenting with cutaneous lesions of BCLs at Memorial Sloan Kettering Cancer Center and Stanford University during 1997-2016. RESULTS: At initial staging, of the 522 patients, extracutaneous disease was noted in 3.6% and 8.8% of patients with marginal zone lymphoma (MZL, n = 306) and follicle center lymphoma (FCL, n = 216) histology, respectively. In patients with systemic involvement, imaging alone identified 81.8% (9/11) of MZL cases and 89.4% of follicular lymphoma cases. In primary cutaneous MZL, 1.7% of patients subsequently had extracutaneous involvement (median follow-up 45 months), and in primary cutaneous FCL. 3.0% subsequently had extracutaneous involvement (median follow-up 47 months). LIMITATIONS: This was a retrospective study. CONCLUSION: Imaging is effective at identifying patients with systemic involvement in indolent BCLs present in the skin; however, incidence is low. After negative initial staging, primary cutaneous MZL patients may be followed clinically without routine imaging.