Skin injury data expansion based on improved generative adversarial network

Ya Gao; Jianan She; Yanzhe Hu; Yu Li

doi:10.1117/12.2680940

8 June 2023 Skin injury data expansion based on improved generative adversarial network

Ya Gao, Jianan She, Yanzhe Hu, Yu Li

Proceedings Volume 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023); 127074S (2023) https://doi.org/10.1117/12.2680940
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 2023, Changsha, China

Abstract

Early detection of skin cancer is the key to increase the chance of successful treatment. Deep learning model has significant effect in automatic classification of skin lesions, but deep learning model requires a large number of data sets, and the availability of annotated images in the medical field is more limited. Data expansion can expand the training data set by transforming the input image. For medical images, the traditional data expansion can only be done by cutting, rotating and other simple operations that do not change the original image features. Therefore, this paper uses generative adjunctive network (GAN) to synthesize medical images of skin lesions for data expansion, so as to obtain more high-quality images with richer content and improve the efficiency of data expansion accordingly. In this paper, the channel focus (ECA) module is added to DCGAN model to improve model performance. Avoid shrinking channel dimensions and reducing the number of references when learning channel attention information. With the addition of a few parameters, the module can obtain obvious performance gain and make the expanded image clearer. The nature of the data allows us to compare directly with the expanded image. Finally, the classification accuracy of the proposed method in ResNet reaches 99.5%, which is 1% and 15.4% higher than that before improvement and without expansion.

Citation Download Citation

Ya Gao, Jianan She, Yanzhe Hu, and Yu Li "Skin injury data expansion based on improved generative adversarial network", Proc. SPIE 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 127074S (8 June 2023); https://doi.org/10.1117/12.2680940

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
9 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Education and training

Data modeling

Skin

Gallium nitride

Image classification

Medical imaging

Image resolution

Show All Keywords

Keywords/Phrases

Search In:

Publication Years