Brain bleeds refer to any condition which results in blood within the skull but outside of its normal location in an artery or vein. When bleeding is from an artery, it is commonly due to an abnormal weakening of the artery wall, such as seen in the case of a ruptured aneurysm. An estimated 6 million people in the United States have an unruptured brain aneurysm. The most common type of venous brain bleeds, on the other hand, is a subdural vein rupture. Brain bleeds can also arise from within the tiny vessels in the brain tissue itself: these are known as “parenchymal” brain bleeds and are most commonly due to stroke, trauma or cancer.
The presence of a bleed within the brain is a finding of major importance which should trigger immediate therapeutic intervention. Unfortunately there are several factors that contribute to the difficulty in making an accurate and timely diagnosis of bleeds on CT brains; firstly, the vast majority of Brain CT scans are completely normal (Callaghan, Kerber, Pace, Skolarus, & Burke, 2014). In the absence of traumatic head injury, only 13% of head CTs show any significant abnormality and less than 5% show evidence of intracranial blood (Wang & You, 2013). Secondly, many bleeds can be subtle on CT imaging (Yuh et al., 2013).
Zebra has developed an algorithm that automatically detects internal brain bleeds based on standard, non-contrast Brain CTs. This tool can provide early detection of people at high risk of severe brain bleeding events.