This paper takes an exploratory case study approach to analyzing the utility of CNN to determine whether intracranial hemorrhage is present on a head cat scan. The motivations of this paper were to determine whether CNN is fast and accurate enough to be used in a clinical setting as well as the looking at the variability in accuracy when using different CNN algorithms as well as different data sets. The conclusion of this paper is that CNN algorithms are both fast and accurate enough to be used in a clinical setting to assist radiology and non-radiology physicians determine if a hemorrhage is present on a head cat scan but is not accurate enough to replace a radiologist. This paper also concluded that it is preferable to use a pre-trained algorithm (vs non-pre-trained algorithm) when limited by using a small data set and pre-trained algorithms with a greater number of layers are a better choice than using a pre-trained algorithm with a lesser number of layers.