Search

Computer-aided skin cancer classification systems built with deep neural networks usually yield predictions based only on images of skin lesions. Despite presenting promising results, it is possible to achieve higher performance by taking into account patient demographics, which are important clues that human experts consider during skin lesion screening. In this article, we deal with the problem of combining images and metadata features using deep learning models applied to skin cancer classification. We propose the Metadata Processing Block (MetaBlock), a novel algorithm that uses metadata to support data classification by enhancing the most relevant features extracted from the images throughout the classification pipeline. We compared the proposed method with two other combination approaches: the MetaNet and one based on features concatenation. Results obtained for two different skin lesion datasets show that our method improves classification for all tested models and performs better than the other combination approaches in 6 out of 10 scenarios.

Skin cancer is one of the most common types of cancer worldwide. Over the past few years, different approaches have been proposed to deal with automated skin cancer detection. Nonetheless, most of them are based only on dermoscopic images and do not take into account the patient clinical information, an important clue towards clinical diagnosis. In this work, we present an approach to fill this gap. First, we introduce a new dataset composed of clinical images, collected using smartphones, and clinical data related to the patient. Next, we propose a straightforward method that includes an aggregation mechanism in well-known deep learning models to combine features from images and clinical data. Last, we carry out experiments to compare the models' performance with and without using this mechanism. The results present an improvement of approximately 7% in balanced accuracy when the aggregation method is applied. Overall, the impact of clinical data on models' performance is significant and shows the importance of including these features on automated skin cancer detection., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2019 Elsevier Ltd. All rights reserved.)