Objectives: Silent cerebral artery vasospasm in aneurysmal subarachnoid hemorrhage causes serious complications such as cerebral ischemia and death. A transcranial Doppler (TCD) ultrasound system is a noninvasive device that can effectively detect cerebral artery vasospasm as soon as it sets in, even before and in the absence of clinical deterioration. Continuous or even daily TCD monitoring is challenging because of the operator expertise and certification required in the form of a trained sonographer and interpretive experience required in the form of an additionally trained and certified physician to perform these studies. This barrier exists because of a lack of automation for detection (without human intervention) of cerebral artery vasospasm using TCD ultrasound. To overcome this barrier, we present an algorithm that automates detection of cerebral artery vasospasm. Methods: We extracted features such as the energy, energy entropy, zero-crossing rate, spectral centroid, spectral speed, spectral entropy, spectral flux, spectral roll-off, harmonic ratio, chroma, and Mel frequency cepstral coefficients for signal classification. Then we applied principal component analysis to reduce the data dimensionality. Results: All of the chosen features were used for training a decision-tree classifier. The algorithm had high accuracy for cerebral artery vasospasm detection, with overall sensitivity of 87.5% and specificity of 89.74%. Conclusions: The algorithm has the potential for development into a continuous cerebral artery vasospasm monitor.