Abnormal blood artery dilation in the brain is a feature of intracranial aneurysms, which carry serious health consequences because they can rupture. Early and accurate diagnosis of these aneurysms is very important for better patient results as it provides for timely intervention. The cumbersome and interpretatively variable nature of conventional diagnosis techniques, which involve manual examination of CT scans, points to the need for a more practical, standardized approach. In this paper, we introduce a Deep Learning-based CNN specifically trained on a model to recognize the various types of intracranial aneurysms from photographs from images acquired via CT scans. Most certainly, this model becomes the very handy clinical equipment which further enables radiologists to build more sensitive diagnosis by providing it with classifications of aneurysms; may further help in processing faster diagnosis by being nicely implemented into healthcare systems coupled with a significant reduction in number of invasive procedures in support of enhanced patient care.