Detection of colorectal cancer using dilated convolutional network via Raman spectra

Title

Detection of colorectal cancer using dilated convolutional network via Raman spectra

Creator

Tanwar S.; Sivakumar V.; Nataraj C.; Vijaylakshmi S.

Description

Colorectal cancer (CRC) is a leading cause of cancer-related deaths worldwide. Early detection plays a crucial role in improving patient outcomes and reducing mortality rates. In recent years, Raman spectroscopy has emerged as a promising tool for non-invasive cancer detection. This research introduces a new method for identifying colorectal cancer (CRC). It combines Raman spectroscopy, a technique that analyzes the molecular fingerprint of tissues, with a powerful deep learning algorithm called a dilated convolutional network (DCN). By combining these two tools, the researchers aim to improve the accuracy and reliability of diagnosing CRC. Intraoperative diagnostics and pathology need to distinguish tumors from normal tissues. This proposal explores Raman spectroscopy as a new surgical tool for identifying colorectal cancer during surgery. Raman spectroscopy offers a way to directly analyze the makeup of tissue, potentially revealing the presence of cancer. However, surrounding tissue can create background interference, making it difficult to detect the key signal. The authors suggest that high-quality data from Raman spectroscopy combined with advanced deep learning algorithms could be a solution to overcome this challenge. We collect a large Raman spectroscopy dataset from 26 colorectal cancer patients with Raman shifts from 385 to 1545 cm-1. Second, dilated convolutional networks classify colorectal cancer tumour tissues. Following the deep learning model's output, we proceed by visualizing and analyzing the identified fingerprint peaks. Our deep learning algorithm exceeds previous colorectal cancer detection methods with 99.1% accuracy. Colorectal cancer detection using Raman spectra is unique. Our ensemble DCN could classify colorectal tumour and normal tissue Raman spectra. 2024 Author(s).

Source

AIP Conference Proceedings, Vol-3161, No. 1

Date

2024-01-01

Publisher

American Institute of Physics

Coverage

Tanwar S., Manipal University, Jaipur, India; Sivakumar V., Asia Pacific University of Technology and Innovation (APU), Kuala Lumpur, Malaysia; Nataraj C., Asia Pacific University of Technology and Innovation (APU), Kuala Lumpur, Malaysia; Vijaylakshmi S., Christ (Deemed to Be University), Pune, India

Rights

Restricted Access

Relation

ISSN: 0094243X

Format

Online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1063/5.0229492

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85203974247&doi=10.1063%2f5.0229492&partnerID=40&md5=7d410e9b1f3ef13ab84367ecd012ea3d