Data augmentation strategies for Raman spectral analysis: application for skin cancer discrimination

Jianhua Zhao; Harvey Lui; Sunil Kalia; Tim K. Lee; Haishan Zeng

doi:10.1117/12.3009005

13 March 2024 Data augmentation strategies for Raman spectral analysis: application for skin cancer discrimination

Jianhua Zhao, Harvey Lui, Sunil Kalia, Tim K. Lee, Haishan Zeng

Author Affiliations +

Proceedings Volume PC12816, Photonics in Dermatology and Plastic Surgery 2024; PC128160F (2024) https://doi.org/10.1117/12.3009005
Event: SPIE BiOS, 2024, San Francisco, California, United States

Abstract

Raman spectroscopy has been evaluated for skin cancer detection. Data augmentation has been used for image processing by deep neural networks. In this study, we proposed and evaluated different data augmentation strategies for spectral augmentation, including added random noise, spectral shift, spectral combination and artificially synthesized spectra using one-dimensional generative adversarial networks (1D-GAN). The stratified samples (n=731) were divided randomly into training (70%), validation (10%) and test dataset (20%), and were repeated 56 times in parallel computing. It was found that data augmentation is not only applicable to deep neural networks, but also applicable to conventional machine learning techniques. When all the strategies were combined to augment the training dataset, the performance of the test dataset could be improved by 2-71%.

Conference Presentation

Citation Download Citation

Jianhua Zhao, Harvey Lui, Sunil Kalia, Tim K. Lee, and Haishan Zeng "Data augmentation strategies for Raman spectral analysis: application for skin cancer discrimination", Proc. SPIE PC12816, Photonics in Dermatology and Plastic Surgery 2024, PC128160F (13 March 2024); https://doi.org/10.1117/12.3009005

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
PRESENTATION

WATCH
PRESENTATION SAVE TO MY LIBRARY

GET CITATION

KEYWORDS

Raman spectroscopy

Skin cancer

Cancer detection

Education and training

Lawrencium

Machine learning

Convolutional neural networks

Show All Keywords

Keywords/Phrases

Search In:

Publication Years