Prediction of Treatment Recommendations Via Ensemble Machine Learning Algorithms for Non-Small Cell Lung Cancer Patients in Personalized Medicine

Aug 5, 2026

Upload by Skyler

This research develops treatment-related genomic predictive markers for non-small cell lung cancer, integrating multiple machine learning approaches to recommend near-optimal individualized chemotherapy. The study compiles a comprehensive meta-database from NCBI Gene Expression Omnibus to capture complex genomic patterns linked to treatment outcomes. Ensemble learning combines bagging with regularized Cox regression and nonparametric tree-based models via Random Survival Forests to support refined, personalized clinical decision-making, addressing patient heterogeneity while improving chemotherapy precision and balancing efficacy versus toxicity risks.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Abstract

Objectives

Methods

Results

Conclusion

Introduction

Prediction of Treatment Recommendations Via Ensemble Machine Learning Algorithms for Non-Small Cell Lung Cancer Patients in Personalized Medicine Hojin Moon1 , Lauren Tran2 , Andrew Lee3 , Taeksoo Kwon4 and Minho Lee5 1Department of Mathematics and Statistics, California State University, Long Beach, Long Beach, CA, USA. 2 Department of Epidemiology, School of Public Health, University of California, Los Angeles, Los Angeles, CA, USA. 3College of Chemistry, University of California, Berkeley, CA, USA. 4School of Information and Computer Science, University of California, Irvine, CA, USA. 5School of Math and Computer Science, Irvine Valley College, Irvine, CA, USA. Cancer Informatics Volume 23: 1–13 © The Author(s) 2024 Article reuse guidelines: [sagepub.com/journals-permissions](sagepub.com/journals-permissions)[ ](sagepub.com/journals-permissions)[DOI: 10.1177/11769351241272397](DOI: 10.1177/11769351241272397) ABSTRACT OBjECTiVES: The primary goal of this research is to develop treatment-related genomic predictive markers for non-small cell lung cancer by integrating various machine learning algorithms that recommends near-optimal individualized patient treatment for chemotherapy in an effort to maximize efficacy or minimize treatment-related toxicity. This research can contribute toward developing a more refined, accurate and effective therapy accounting for specific patient needs . METhOdS: To accomplish our research goal, we implement ensemble learning algorithms, bagging with regularized Cox regression models and nonparametric tree-based models via Random Survival Forests . A comprehensive meta-database was compiled from the NCBI Gene Expression Omnibus data repository for lung cancer patients to capture and utilize complex genomic patterns that can predict treatment outcomes more accurately. RESuLTS: The developed novel prediction algorithm demonstrates the ability to support complex clinical decision-making processes in the treatment of NSCLC. It effectively addresses patient heterogeneity, offering predictions that are both refined and personalized in improving the precision of chemotherapy regimens prescribed to the eligible patients . CONCLuSiON: This research should contribute substantial advancement of cancer treatments by improving the accuracy and efficacy of chemotherapy treatments for a targeted group of patients who need the right treatment. The integration of complex machine learning techniques with genomic data holds substantial potential to transform current cancer treatment paradigms by providing robust support in clinical decision-making. KEywORdS: Biomedical data science, cancer genomics, genomic biomarkers, personalized chemotherapy, precision oncology RECEiVEd: January 28, 2024. ACCEPTEd: July 14, 2024. TyPE:Original Research FuNdiNg: The author(s) received no financial support for the research, authorship, and/or publication of this article. dECLARATiON OF CONFLiCTiNg iNTERESTS: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. CORRESPONdiNg AuThOR: Hojin Moon, Department of Mathematics and Statistics, California State University, Long Beach, 1250 N Bellflower Blvd. , Long Beach, CA 90840-1001, USA. Email: [hojin.moon@csulb.edu](hojin.moon@csulb.edu) | Introduction | patients with ACT carefully such that chances of success are | | --- | --- | | Non-small-cell lung cancer (NSCLC) affects over 200 000 | high enough to justify the risk of relapse and metastatic | | Americans per year and it constitutes 85% of lung cancer.1 | potential. | | Physicians consider the best medical treatment options to | Advancements in biotechnology in recent years have | | improve quality of life and prolong survival in patients with | increased the availability of high-dimensional genomic data for | | NSCLC. The best treatment option for early-stage NSCLC is | biomedical decision making. For such data t

FAQs

What is the main goal of the study on NSCLC treatment recommendations?

To develop genomic predictive markers that integrate machine learning algorithms to recommend near-optimal individualized chemotherapy for non-small cell lung cancer patients, aiming to maximize efficacy or minimize toxicity.

Which ensemble machine learning methods are used?

The study implements ensemble learning combining bagging with regularized Cox regression models and nonparametric tree-based models through Random Survival Forests.

How is the training data prepared for predicting treatment outcomes?

A comprehensive meta-database is compiled from the NCBI Gene Expression Omnibus repository to capture complex genomic patterns associated with lung cancer patients and treatment outcomes.

Aug 5, 2026

Upload by Skyler