Accidental brain injuries are the major cause of haemorrhages, which can be difficult to detect due to the similar intensity and disorientation of brain- related diseases like haemorrhages. Computer-assisted disease diagnosis (CAD) has the potential to detect and identify brain-related diseases. Computed tomography and magnetic resonant imaging-like modalities are sensitive to haemorrhage detection, and a novel study has revealed complexity in the modulation of vascular and bone haemorrhage. Because of the vast amount of medical images generated each day, the discipline of medical imaging is focused on inventing equipment to aid medical professionals. Because of distortions and extraneous information, brain imaging might be difficult to diagnose bleeding spots in the brain haemorrhage. A deep learning strategy under AI that can efficiently diagnose and detect brain-related diseases with multi-modality fusion, as opposed to the traditional method, which is tedious and time-consuming. The analysis of brain hemorrhage using DL is indispensable to help medical professionals address these illnesses. These concerns must be identified for medical diagnosis, treatment, and research. This paper describes the previously published review papers in brain-related disease diagnosis, AI approaches, multi- modality, challenges, and future work.