ABSTRACT
Traumatic brain injury can occur because of a violent blow or jolt to the head that can result in long-term complications or death. Automated methods can facilitate quantifying the amount of lesions such as hematoma and edema, which could be used in prognostic models and hence help clinicians in their decision making. In this paper, we review state-of-the-art methods developed for analyzing head CT images for identifying/quantifying hematoma, edema and midline shift, and also predictive methods developed for outcomes prediction. We categorize these methods based on their approaches and provide details on their evaluation metrics and validation techniques.
