A feature fusion system for basal cell carcinoma detection through data‐driven feature learning and patient profile

Abstract

Background: Basal cell carcinoma (BCC) is the most common skin cancer, which is highly damaging in its advanced stages. Computer-aided techniques provide a feasible option for early detection of BCC. However, automated BCC detection techniques immensely rely on handcrafting high-level precise features. Such features are not only computationally complex to design but can also represent a very limited aspect of the lesion characteristics. This paper proposes an automated BCC detection technique that directly learns the features from image data, eliminating the need for handcrafted feature design. Methods:The proposed method is composed of 2 parts. First, an unsupervised feature learning framework is proposed which attempts to learn hidden characteristics of the data including vascular patterns directly from the images. This is done through the design of a sparse autoencoder (SAE). After the unsupervised learning, we treat each of the learned kernel weights of the SAE as a filter. Convolving each filter with the lesion image yields a feature map. Feature maps are condensed to reduce the dimensionality and are further integrated with patient profile information. The overall features are then fed into a softmax classifier for BCC classification. Results: On a set of 1199 BCC images, the proposed framework achieved an area under the curve of 91.1%, while the visualization of learned features confirmed meaningful clinical interpretation of the features. Conclusion: The proposed framework provides a non-invasive fast BCC detection tool that incorporates both dermoscopic lesional features and clinical patient information, without the need for complex handcrafted feature extraction. K E Y W O R D S automated basal cell carcinoma detection, blood vessels, dermoscopy, feature learning, sparse autoencoders

Multimodal skin lesion classification using deep learning

“…There were early efforts to apply neural networks to dermatoscopic skin lesion classification; however, the subsequent years focused mainly on image processing techniques and techniques for feature extraction . In recent years, there has been a shift back towards the end‐to‐end application of neural networks . This is largely thanks to an exponential increase in GPU computing capability as well as an overall improvement in the effectiveness of convolutional neural networks (both through significant research in network design and the curation of large data sets such as ImageNet).…”

“…The focus of this work is to explore the importance of the dermatoscopic imaging modality specifically in conjunction with its macroscopic counterpart for the task of automated lesion diagnosis. We also include comparisons with previous studies that leverage patient‐level metadata as this has been shown to improve diagnostic accuracy . Our network architecture, shown in Figure , is chosen using a grid‐search technique while trying to maintain overall simplicity where possible.…”

The impact of patient clinical information on automated skin cancer detection

“…In this sense, Brinker et al [42] presented a review for deep learning models applied to skin cancer detection and concluded that an improvement in classification quality could be achieved by adding clinical data in the classification process. Based on this idea, Kharazmi et al [43] proposed a deep learning approach to detect BCC using dermoscopic images and five patients clinical information. The results using the clinical data present an improvement, however, they did not analysis how much it affect in the classification.…”

Artificial intelligence in dermatology: the ‘unsupervised’ learning

“…The meaning of these clusters can be determined later, such as ‘malignant lesions’ and ‘benign lesions’. Kharazmi et al ., in their study of basal cell carcinoma (BCC) detection, used an unsupervised feature learning framework to achieve an area under the curve of 91·1% . Their study integrated patient information along with dermoscopic images of BCCs to achieve the result.…”