Skin Lesion Classification On Imbalanced Data Using Deep Learning With Soft Attention

Today, the rapid development of industrial zones leads to an increased incidence of skin diseases because of polluted air. According to a report by the American Cancer Society, it is estimated that in 2022 there will be about 100,000 people suffering from skin cancer and more than 7600 of these people will not survive. In the context that doctors at provincial hospitals and health facilities are overloaded, doctors at lower levels lack experience, and having a tool to support doctors in the process of diagnosing skin diseases quickly and accurately is essential. Along with the strong development of artificial intelligence technologies, many solutions to support the diagnosis of skin diseases have been researched and developed. In this paper, a combination of one Deep Learning model (DenseNet, InceptionNet, ResNet, etc) with Soft-Attention, which unsupervisedly extracts a heat map of main skin lesions. Furthermore, personal information including age and gender is also used. It is worth noting that a new loss function that takes into account the data imbalance is also proposed. Experimental results on data set HAM10000 show that using InceptionResNetV2 with Soft-Attention and new loss function give 90 percent accuracy, mean of precision, f1-score, recall- score, and AUC scores of 0.81, 0.81, 0.82, and 0.99, respectively. Besides, using MobileNetV3Large 14 combined with Soft-Attention and new loss function, even though the number of parameters is 11 15 times less, the number of hidden layers is 4 times less, it achieves an accuracy of 0.86 and 30 times 16 faster diagnosis than InceptionResNetV2.

Keywords: Skin Lesions, Classification, Deep Learning, Soft-Attention, Imbalance

Citation

@article{
  title={Soft-Attention Improves Skin Cancer Classification Performance},
  author={Viet Dung Nguyen, Ngoc Dung Nguyen, and Hoang Khoi Do},
  journal={Sensors MDPI},
  year={2022}
}

Results

ROC of InceptionResNetV2 and DenseNet on HAM10000

Datasets

HAM10000 dataset:

Tschandl, P., Rosendahl, C. & Kittler, H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci. Data 5, 180161 doi:10.1038/sdata.2018.161 (2018).Available: https://www.nature.com/articles/sdata2018161, https://arxiv.org/abs/1803.10417

ISIC-2017 dataset:

Codella N, Gutman D, Celebi ME, Helba B, Marchetti MA, Dusza S, Kalloo A, Liopyris K, Mishra N, Kittler H, Halpern A. "Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC)". arXiv: 1710.05006 [cs.CV] Available: https://arxiv.org/abs/1710.05006