Cancer-Net SCa: tailored deep neural network designs for detection of skin cancer from dermoscopy images

Abstract

Background: Skin cancer continues to be the most frequently diagnosed form of cancer in the U.S., with not only significant effects on health and well-being but also significant economic costs associated with treatment. A crucial step to the treatment and management of skin cancer is effective early detection with key screening approaches such as dermoscopy examinations, leading to stronger recovery prognoses. Motivated by the advances of deep learning and inspired by the open source initiatives in the research community, in this study we introduce Cancer-Net SCa, a suite of deep neural network designs tailored for the detection of skin cancer from dermoscopy images that is open source and available to the general public. To the best of the authors' knowledge, Cancer-Net SCa comprises the first machine-driven design of deep neural network architectures tailored specifically for skin cancer detection, one of which leverages attention condensers for an efficient self-attention design.

Results: We investigate and audit the behaviour of Cancer-Net SCa in a responsible and transparent manner through explainability-driven performance validation. All the proposed designs achieved improved accuracy when compared to the ResNet-50 architecture while also achieving significantly reduced architectural and computational complexity. In addition, when evaluating the decision making process of the networks, it can be seen that diagnostically relevant critical factors are leveraged rather than irrelevant visual indicators and imaging artifacts.

Conclusion: The proposed Cancer-Net SCa designs achieve strong skin cancer detection performance on the International Skin Imaging Collaboration (ISIC) dataset, while providing a strong balance between computation and architectural efficiency and accuracy. While Cancer-Net SCa is not a production-ready screening solution, the hope is that the release of Cancer-Net SCa in open source, open access form will encourage researchers, clinicians, and citizen data scientists alike to leverage and build upon them.

Keywords: Akin cancer; Deep neural network; Melanoma; Self-attention.

Fig. 1

Sample images from the ISIC Dataset leveraged to build Cancer-Net SCa. Dermoscopy images a and b are both benign, while images c and d are both malignant. Image b can easily be mistaken for a malignant lesion, while image c can easily be misclassified as benign to the untrained eye

Fig. 2

The proposed Cancer-Net SCa network architectures. The number in each convolution module represents the number of channels. The numbers in each visual attention condenser represents the number of channels for the down-mixing layer, the embedding structure, and the up-mixing layer, respectively (details can be found in [15]). It can be observed that all Cancer-Net SCa architectures exhibit both great macroarchitecture and microarchitecture design diversity, with certain models exhibiting specific lightweight macroarchitecture design characteristics such as attention condenser and projection–expansion–projection–expansion (PEPE) design patterns comprised of channel dimensionality reduction, depthwise convolutions, and pointwise convolutions

Fig. 3

Sample images from the ISIC Dataset that the trained models were tested on. All seven tested model architectures correctly identified image a as malignant. However, the DenseNet and Inception architectures incorrectly classified image b as benign, while all three Cancer-Net SCa models had a correct classification of malignant

Fig. 4

Example dermoscopy images of malignant cases from the ISIC dataset and their associated diagnostically relevant imaging features as identified by GSInquire [56], using Cancer-Net SCa-A. The bright regions indicate the imaging features identified to be relevant

Fig. 5

Example dermoscopy images of benign cases from the ISIC dataset and their associated diagnostically relevant imaging features as identified by GSInquire [56], using Cancer-Net SCa-A. The bright regions indicate the imaging features identified to be relevant