Skin Lesion Segmentation Based on Multi-Scale Attention Convolutional Neural Network
| Autor: | Simin Cao, Hai Zhang, Yun Jiang, Shengxin Tao |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
General Computer Science
Channel (digital image) Computer science multi-scale Pooling 02 engineering and technology Residual Convolutional neural network 030218 nuclear medicine & medical imaging Lesion 03 medical and health sciences 0302 clinical medicine 0202 electrical engineering electronic engineering information engineering medicine General Materials Science Segmentation Pyramid (image processing) Block (data storage) business.industry skin lesion segmentation General Engineering Pattern recognition 020201 artificial intelligence & image processing Artificial intelligence lcsh:Electrical engineering. Electronics. Nuclear engineering medicine.symptom Deep convolutional neural network business attention mechanism lcsh:TK1-9971 |
| Zdroj: | IEEE Access, Vol 8, Pp 122811-122825 (2020) |
| ISSN: | 2169-3536 |
| Popis: | The incidence of skin cancer around the world is increasing year by year. However, early diagnosis and treatment can greatly improve the survival rate of patients. Skin lesion boundary segmentation is essential to accurately locate lesion areas in dermatoscopic images. It is true that accurate segmentation of skin lesions is still challenging dues to problems such as blurred borders, which requires an accurate and automatic skin lesion segmentation method. In this paper, we propose an end-to-end framework which can perform skin lesion segmentation automatically and efficiently, called the CSARM-CNN (Channel & Spatial Attention Residual Module) model. Each CSARM block of the model combines channel attention and spatial attention to form a new attention module to enhance segmentation results. The multi-scale input images are obtained by the spatial pyramid pooling. Finally, a weighted cross-entropy loss function is used at each side of the output layer to sum the total loss of the model. We evaluated in two published standard datasets, ISIC 2017 and PH2, and achieved competitive results in terms of specificity and accuracy, with 99.03% and 99.45% specificity, 94.96% and 95.23% accuracy, respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|