Abstract
Tensor decomposition has emerged as a powerful framework for featureextraction in multi-modal biomedical data. In this review, we present acomprehensive analysis of tensor decomposition methods such as Tucker,CANDECOMP/PARAFAC, spiked tensor decomposition, etc. and their diverseapplications across biomedical domains such as imaging, multi-omics, andspatial transcriptomics. To systematically investigate the literature, weapplied a topic modeling-based approach that identifies and groups distinctthematic sub-areas in biomedicine where tensor decomposition has been used,thereby revealing key trends and research directions. We evaluated challengesrelated to the scalability of latent spaces along with obtaining the optimalrank of the tensor, which often hinder the extraction of meaningful featuresfrom increasingly large and complex datasets. Additionally, we discuss recentadvances in quantum algorithms for tensor decomposition, exploring how quantumcomputing can be leveraged to address these challenges. Our study includes apreliminary resource estimation analysis for quantum computing platforms andexamines the feasibility of implementing quantum-enhanced tensor decompositionmethods on near-term quantum devices. Collectively, this review not onlysynthesizes current applications and challenges of tensor decomposition inbiomedical analyses but also outlines promising quantum computing strategies toenhance its impact on deriving actionable insights from complex biomedicaldata.