# Cancer survival prediction based on soft-label guided contrastive learning and global feature fusion

## TL;DR

This paper introduces a new cancer survival prediction model that uses advanced machine learning techniques to better integrate multi-omics data.

## Contribution

The novel model, SLCGF, uses soft-label guided contrastive learning and global feature fusion for improved cancer survival prediction.

## Key findings

- SLCGF outperforms existing methods in cancer survival prediction on 13 multi-omics datasets.
- The model effectively integrates unique and shared information from different omics data.
- Soft-label guidance enhances feature discrimination and alignment across omics.

## Abstract

The high complexity and heterogeneity of cancer pose significant challenges to personalized treatment, making the improvement of cancer survival prediction accuracy crucial for clinical decision-making. The integration of multi-omics data enables a more comprehensive capture of multi-layered information in complex biological processes. However, existing survival analysis models still face limitations in accurately extracting and effectively integrating the unique and shared information from multi-omics data.

In this article, we propose a novel prediction model for cancer survival based on soft-label guided contrastive learning and global feature fusion, namely SLCGF. Our model first extracts paired feature representations for each omics using Siamese encoders. We then perform intra-view and inter-view contrastive learning simultaneously, employing a neighborhood-based paradigm to enhance feature discrimination and alignment across omics. To ensure reliable neighbor retention and improve model robustness, we treat the affinities between samples and their high-order neighbors as soft labels to guide the contrastive learning process at both levels. In addition, we adopt a global self-attention mechanism to obtain the unified representation for cancer survival prediction, where the cross-omics connections are fully exploited and complementary information is adaptively integrated. We comprehensively evaluate the performance of our model on 13 cancer multi-omics datasets, and the experimental results demonstrate its superiority over existing approaches.

Source code is available at https://github.com/LiangSDNULab/SLCGF.