A multi-modal fusion framework based on multi-task correlation learning for cancer prognosis prediction

作者：

Highlights：

• We propose that not only multiple modalities, but also multiple tasks can improve the performance of cancer diagnosis and prognosis prediction.

• We use sparse graph convolutional network (SGCN) to learn the representations of mRNA expression data via gene-gene interactions.

• Experiment results demonstrate that multi-task learning can improve the performance of multiple tasks simultaneously.

摘要

•We propose that not only multiple modalities, but also multiple tasks can improve the performance of cancer diagnosis and prognosis prediction.•We use sparse graph convolutional network (SGCN) to learn the representations of mRNA expression data via gene-gene interactions.•Experiment results demonstrate that multi-task learning can improve the performance of multiple tasks simultaneously.

论文关键词：Multi-modal fusion,Multi-task learning,Survival analysis,Cancer grade

论文评审过程：Received 11 March 2021, Revised 7 January 2022, Accepted 16 February 2022, Available online 24 February 2022, Version of Record 26 February 2022.

论文官网地址：https://doi.org/10.1016/j.artmed.2022.102260