Subarachnoid Hemorrhage

Value Proposition

Subarachnoid hemorrhage (SAH) is typically either post-traumatic or the result of a ruptured intracranial aneurysm or vascular malformation. It can occur in patients of any demographic. SAH is critical to identify, since hemorrhage in the subarachnoid space can predispose the patient to cerebral ischemia and hydrocephalus. In the setting of a recently ruptured intracranial aneurysm or vascular malformation, an initial “sentinel hemorrhage” may be followed by a devastating second hemorrhage that may lead to sudden death. Identifying SAH as soon as possible can significantly improve patient outcomes. In emergency situations, automated identification and classification of SAH would assist case prioritization and interpretation. In busy private practices, outpatient CT scans that potentially demonstrate SAH may not be reviewed for days after completion, delaying management and potentially leading to poor patient outcomes. Automated identification and classification would also provide value to non-specialists, especially those who read images in emergency situations.

Narrative

A 28-year-old woman experiences the “worst headache of her life”. She takes numerous acetaminophen tablets and drives herself to the nearest community hospital emergency room. The emergency room physician examines her and notices that her pupils are unequal in size. Her headache persists but has slightly improved. A non-contrast CT scan of her head is ordered.

Workflow Description

Patient receives a CT of their head; the clinical indication will usually be head trauma or headache. Algorithm receives entire CT dataset. If the algorithm can determine a result, return: density, volume, location (R/L; frontal/parietal/temporal/occipital/basilar cisterns/cerebellar). A reliability metric, assessing the accuracy of the volumetric measurement, would also be helpful. Following identification of the SAH volume, location and density, the algorithm should give a probability that the hemorrhage is post-traumatic, aneurysmal, or other. An alert should notify the user.
Additional considerations are as follows: Algorithm executes after exam is verified on PACS in the backend. Algorithm optimally integrates on PACS and dictation or reporting software. The user is able to automatically populate the report or manually input the results. Indicator image may save to PACS as part of the medical record.

Inputs

DICOM Study

Primary Outputs

SAH Laterality

SAH Location

SAH Volume

SAH Density 

SAH Etiology