Tensor-based Fusion of Multi-modal Data for Predicting Cancer Risk Levels
Abstract
Cancer diagnosis and treatment increasingly rely on multi-modal diagnostic technologies that generate high-dimensional datasets. Although these datasets are rich in pathological information, they pose considerable challenges in interpretation, dimensionality reduction, and clinical applicability. Conventional machine learning approaches often fail to extract meaningful cross-modal features, thereby limiting predictive accuracy in patient risk stratification. To address these limitations, this paper proposes an unsupervised machine learning framework that integrates feature extraction, tensor decomposition, and clustering to improve cancer risk prediction. The framework employs tensor decomposition to fuse heterogeneous genomic and clinical data, enabling the extraction of latent, biologically meaningful features that outperform unimodal feature sets. Multiple feature extraction strategies are investigated, and the resulting features are subsequently clustered using a range of clustering algorithms. Comparative evaluations based on internal and external metrics are then used to identify the most effective configurations. The framework is validated on LinkedOmics data comprising 11,158 patients across 32 cancer types. For patient risk stratification, clinically relevant attributes such as tumor purity, pathological stage, and survival outcomes serve as ground-truth indicators. Experimental results demonstrate the superiority of tSVD+GMM, which consistently outperforms other configurations across clinical benchmarks.
Related
- MF-TLMLP: A multi-channel fusion-based multi-fidelity transfer learning method for predicting the cutting performance of conical picks
- Adaptive gating-based fusion of sequential and local features for predicting drug-drug interactions
- Multi-omics analysis reveals protein-driven modulation of growth and metabolism in Gangba sheep under different dietary protein levels
- MSCF-DM: center-site-aware attention and gated multi-scale fusion network for cross-species DNA methylation site prediction
- Fusion of Multi-Spectral Features from UAVs to Improve the Accuracy of Total Soil Nitrogen Estimation During the Corn Growing Season
BibTeX
@misc{anaissi2026tensorbased,
title = {{Tensor-based Fusion of Multi-modal Data for Predicting Cancer Risk Levels}},
author = {Ali Anaissi and Yuanzhe Jia and Qishun Peng and Weidong Huang and Ali Braytee and Junaid Akram},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.13221},
eprint = {2610.13221},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.13221},
}
RIS
TY - UNPB
TI - Tensor-based Fusion of Multi-modal Data for Predicting Cancer Risk Levels
AU - Ali Anaissi
AU - Yuanzhe Jia
AU - Qishun Peng
AU - Weidong Huang
AU - Ali Braytee
AU - Junaid Akram
PY - 2026
DA - 2026/10/09
PB - arXiv
DO - 10.48550/arxiv.2610.13221
UR - https://arxiv.org/abs/2610.13221
AB - Cancer diagnosis and treatment increasingly rely on multi-modal diagnostic technologies that generate high-dimensional datasets. Although these datasets are rich in pathological information, they pose considerable challenges in interpretation, dimensionality reduction, and clinical applicability. Conventional machine learning approaches often fail to extract meaningful cross-modal features, thereby limiting predictive accuracy in patient risk stratification. To address these limitations, this paper proposes an unsupervised machine learning framework that integrates feature extraction, tensor decomposition, and clustering to improve cancer risk prediction. The framework employs tensor decomposition to fuse heterogeneous genomic and clinical data, enabling the extraction of latent, biologically meaningful features that outperform unimodal feature sets. Multiple feature extraction strategies are investigated, and the resulting features are subsequently clustered using a range of clustering algorithms. Comparative evaluations based on internal and external metrics are then used to identify the most effective configurations. The framework is validated on LinkedOmics data comprising 11,158 patients across 32 cancer types. For patient risk stratification, clinically relevant attributes such as tumor purity, pathological stage, and survival outcomes serve as ground-truth indicators. Experimental results demonstrate the superiority of tSVD+GMM, which consistently outperforms other configurations across clinical benchmarks.
ER -