Intracerebral Hemorrhage (ICH) is the deadliest subtype of stroke, necessitating timely and accurate prognostic evaluation to reduce mortality and disability. However, the multi-factorial nature and complexity of ICH make methods based solely on computed tomography (CT) image features inadequate. Despite the capacity of cross-modal networks to fuse additional information, the effective combination of different modal features remains a significant challenge. In this study, we propose a joint-attention fusion-based 3D cross-modal network termed ICHPro that simulates the ICH prognosis interpretation process utilized by neurosurgeons. ICHPro includes a joint-attention fusion module to fuse features from CT images with demographic and clinical textual data. To enhance the representation of cross-modal features, we introduce a joint loss function. ICHPro facilitates the extraction of richer cross-modal features, thereby improving classification performance. Upon testing our method using a five-fold cross-validation, we achieved an accuracy of 89.11%, an F1 score of 0.8767, and an AUC value of 0.9429. These results outperform those obtained from other advanced methods based on the test dataset, thereby demonstrating the superior efficacy of ICHPro. The code is available at our Github: https://github.com/YU-deep/ICH.