Tumor classification using microRNA expression features and machine learning

Yucong Yan

doi:10.1117/12.2660286

6 December 2022 Tumor classification using microRNA expression features and machine learning

Yucong Yan

Proceedings Volume 12458, International Conference on Biomedical and Intelligent Systems (IC-BIS 2022); 124583B (2022) https://doi.org/10.1117/12.2660286
Event: International Conference on Biomedical and Intelligent Systems, 2022, Chengdu, China

Abstract

Tumor classification and diagnosis are not an easy task as novel research keeps emerging about what makes a tumor type unique. There are several ways to classify a tumor, such as molecular analysis as well as the usual ways where biopsies or physical examinations are used to diagnose cancer. Previous studies have been using DNA or messenger RNA features in tumor classification, but there are not a lot of specific studies in the area of tumor classification using microRNA (miRNA). As new investigations suggest, miRNA serves a crucial role in tumor growth and metastasis, so my investigation dives into the relationship between miRNA and tumor classification, which is still mostly unexplored. Here, using miRNA expression data from large cancer sequencing projects for over 10,000 patients, I build machine learning models to classify more than 30 different types of cancer. Two different algorithms, k-nearest neighbors and random forest are tested and benchmarked. For certain cancer types, the random forest model can be over 95% accurate. I further demonstrate that errors of the classifier come largely from tumors with similar cell-of-origin, such as digestive organ tumor stomach cancer are sometimes mislabeled as other digestive organ tumors like esophageal cancer or colon cancer. To explore how we can use the miRNA-based tumor classifier in a clinical setting, I further demonstrate that it is possible to reach near 95% accuracy with a minimal model of only 10 miRNA genes, which suggests that a small test kit for 10 miRNA genes can reach a comparable accuracy as the more expensive option of whole miRNA sequencing. Overall, this study suggests that miRNA expression can accurately predict tumor types and may be developed into clinical diagnostic kits.

Citation Download Citation

Yucong Yan "Tumor classification using microRNA expression features and machine learning", Proc. SPIE 12458, International Conference on Biomedical and Intelligent Systems (IC-BIS 2022), 124583B (6 December 2022); https://doi.org/10.1117/12.2660286

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

Tumor growth modeling

Cancer

Tumors

Data modeling

Machine learning

Colorectal cancer

Statistical analysis

Show All Keywords

Keywords/Phrases

Search In:

Publication Years