AI-Driven Multi-Omics Integration for Early Disease Detection: A Comprehensive Survey
TL;DR
By comparing genomic and multi-omics strategies, this survey highlights ongoing hurdles related to privacy, bias, interpretability, and scalability.
Abstract
artificial intelligence (AI) is transforming the way we stumble on ailments early, but relying on a single form of facts—which includes genomics on my own—offers most effective a restrained picture of the whole organic complexity. Multi-omics integration—alongside aspect genomics, transcriptomics, proteomics, metabolomics, microbiome information, and scientific signs and symptoms and signs and symptoms—offers a more whole view of illness improvement. modern-day-day research show that graph neural networks (GNNs), federated getting to know (FL), and explainable AI (XAI) outperform genomic-most effective models through identifying novel biomarkers and enhancing diagnostic accuracy. Examples embody Tab net fusion for Alzheimer’s, multimodal deep reading for rheumatoid arthritis, and semi-supervised analyzing for hepatocellular carcinoma. By comparing genomic and multi-omics strategies, this survey highlights ongoing hurdles related to privacy, bias, interpretability, and scalability. destiny tips which consist of transformer-based fusion and basis fashions promise equitable, transparent, and clinically relevant AI-driven multi-omics structures for precision medicine.