INTERNATIONAL CONGRESS OF
PharmacoGenetics
AI-driven Multi-Omics Integration for Personalized Cancer Risk Prediction and Prevention
-
Hamidreza Alimirzaei,1,*
1. Trainee at Atieh Hospital Laboratory, Hamedan, Iran
- Introduction: Cancer continues to be one of the most significant global health challenges, imposing a substantial burden on healthcare systems and contributing to increased economic costs. Early prediction and prevention of cancer can dramatically reduce mortality rates associated with the disease. Recent advancements in molecular sciences and innovative techniques such as artificial intelligence (AI) have made it possible to integrate multi-omics data (including genomics, proteomics, metabolomics, microbiomics, and radiomics) for cancer risk prediction and the development of personalized prevention strategies. This systematic review explores and synthesizes recent research in this area, emphasizing the importance of combining diverse omics data and AI models to improve cancer prediction accuracy.
- Methods: This paper is based on a systematic review approach, gathering articles published in reputable scientific journals between 2020 and 2025 from databases such as PubMed, Scopus, and IEEE Xplore. Selected articles were assessed for quality and credibility, and only the highest standard publications were included. This study integrates genetic, epigenetic, proteomic, metabolomic, microbiomic, and imaging data using advanced AI models, particularly deep learning (DL) and machine learning (ML) models. Convolutional Neural Networks (CNNs) and Transformer models have been applied to process complex omics data and enhance cancer risk prediction accuracy.
- Results: The findings demonstrate that AI models have significantly improved prediction accuracy for various cancers, including breast, colorectal, and lung cancers. These models, by combining genomic data and medical imaging such as CT scans and MRIs, have increased prediction accuracy substantially. Additionally, the integration of microbiome and metabolomics data with genetic and proteomic data has enhanced the models' ability to identify more complex patterns in cancers, leading to higher prediction accuracy. For example, a model that integrated imaging and genetic data for lung cancer prediction achieved an accuracy of 89%. Transformer-based models analyzing microbiomic data have also identified novel associations between microbiota and cancer, further improving prediction capabilities.
- Conclusion: The integration of AI with multi-omics data has not only enhanced cancer prediction accuracy but also enabled the development of personalized prevention strategies for individuals at high risk. These methods can reduce treatment costs and significantly alleviate the economic burden of cancer. However, challenges such as data diversity, access to large datasets, and ethical concerns related to patient privacy remain, and these issues must be addressed to optimize these approaches. In the future, combining these technologies with continuous monitoring tools such as liquid biopsies and wearable devices could play a crucial role in improving cancer prevention and treatment programs.
- Keywords: AI,Genomics,Cancer
