Early esophageal detection using hyperspectral engineering and convolutional neural network

Arvind Mukundan; Yu-Ming Tsao; Hsiang-Chen Wang

doi:10.1117/12.2689086

27 November 2023 Early esophageal detection using hyperspectral engineering and convolutional neural network

Arvind Mukundan, Yu-Ming Tsao, Hsiang-Chen Wang

Proceedings Volume 12770, Optics in Health Care and Biomedical Optics XIII; 1277009 (2023) https://doi.org/10.1117/12.2689086
Event: SPIE/COS Photonics Asia, 2023, Beijing, China

Abstract

The current article introduces an innovative methodology for the categorization and identification of esophageal cancer through the utilization of hyperspectral imaging (HSI) and a deep learning model. The study focuses on identifying the cancer stage and marking its locations within the esophagus. A specialized algorithm has been employed in this study to generate simulated spectrum data from the hyperspectral imaging (HSI) data. This simulated data is then integrated with deep learning methodologies in a unified identification system based on the single-shot multibox detector (SSD). In order to assess the efficacy of the predictive model, dataset comprising of 155 white-light endoscopic (WLI) and 153 narrow-band endoscopic (NBI) images of esophageal cancer has been utilized. The algorithm effectively forecasts the outcomes of 308 examination images in a time frame of 19 seconds. The spectral data-based test results exhibit a precision of 88% for WLI esophageal cancer and 91% for NBI esophageal cancer images. Comparatively, the accuracy using RGB images is 83% for WLI and 86% for NBI. The present research exhibits noteworthy enhancement in prognostication precision, whereby the HSI detection approach elevates the precision of WLI and NBI by 5%. The findings of this research highlight the potential of combining HSI and deep learning for the detection and diagnosis of esophageal cancer. The proposed methodology offers enhanced precision in the identification of cancer stages and the precise demarcation of location. The outcomes of this study present encouraging prospects for advancement and utilization of HSI-derived techniques in the domain of esophageal cancer identification.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Arvind Mukundan, Yu-Ming Tsao, and Hsiang-Chen Wang "Early esophageal detection using hyperspectral engineering and convolutional neural network", Proc. SPIE 12770, Optics in Health Care and Biomedical Optics XIII, 1277009 (27 November 2023); https://doi.org/10.1117/12.2689086

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
8 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Cancer

Endoscopy

RGB color model

Tumor growth modeling

Cancer detection

Education and training

Data modeling

Show All Keywords

Keywords/Phrases

Search In:

Publication Years