Lightweight convolutional neural network model for skin lesions classification

Gang Wang; Peng Qian; Ruirui Tian; Junyi Wang

doi:10.1117/12.3033617

12 July 2024 Lightweight convolutional neural network model for skin lesions classification

Gang Wang, Peng Qian, Ruirui Tian, Junyi Wang

Proceedings Volume 13185, International Conference on Communication, Information, and Digital Technologies (CIDT2024) ; 131850D (2024) https://doi.org/10.1117/12.3033617
Event: International Conference on Communication, Information and Digital Technologies, 2024, Wuhan, China

Abstract

The classification of skin lesion images is challenging because to the substantial intra-class variability and mortality rate of skin cancer. Convolutional neural networks (CNNs) have recently been employed by specialists for dermatology-assisted diagnosis. We provide a lightweight convolutional neural network model for skin lesion classification to enhance the network model's precision in classifying skin lesions. We employ MobileNet-V2 as our backbone network, and the MobileNet-V2 block is built using convolution and an inverted residual block. To enhance the model's ability to depict pathological features, we add an improved interference coordinate attention block (RICA-block) to the inverted residual block. In order to limit the impact of complicated irrelevant backgrounds in lesion images on the classification performance of the model, we developed the RICA-block to perform distinct feature extraction for lesioned and non-lesioned regions of lesion images. The experimental findings demonstrate that, on the test datasets for ISIC2018 and ISIC2019, respectively, our model, which is just 3.32M, achieves classification accuracy of 89.52% and 93.24%, as well as 95% macro averages.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Gang Wang, Peng Qian, Ruirui Tian, and Junyi Wang "Lightweight convolutional neural network model for skin lesions classification", Proc. SPIE 13185, International Conference on Communication, Information, and Digital Technologies (CIDT2024) , 131850D (12 July 2024); https://doi.org/10.1117/12.3033617

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
11 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Skin

Data modeling

Performance modeling

Tumor growth modeling

Visual process modeling

Image classification

Education and training

Show All Keywords

Keywords/Phrases

Search In:

Publication Years