@@@@Highlights •We collected data and established the first multi-omics, multi-modal stroke recurrence dataset.•We proposed the first-ever Integrated Multi-omics Prediction Model for Stroke Recurrence (MPSR).•We introduce the Lnet regularization layer which can effective extraction of salient features from multi-sequence MRI data, concomitantly mitigating noise.•We propose a dynamic weighting mechanism that allows the model to focus on important omics information.•Experimental results show that the MPSR model is superior to the single omics prediction model and the existing multi-omics prediction model.