Algorithm for computer aided diagnosis in spontaneous intracranial hypotension

摘要

Spontaneous Intracranial Hypotension disease is based on tears on the dura mater, which surrounds brain and spinal cord. Since it is a rare disease, and symptoms are so variable, diagnosis is problematic worldwide. Diagnostic lumbar puncture, one of the methods used to diagnose SIH, is an invasive technique in which physician measures opening pressure of cerebrospinal fluid. Though this technique is commonly used for diagnosing SIH, it is invasive and may cause pain, inflammation or new leakages. It is not a certain test as well. Noninvasive MR imaging is used for diagnosis alternatively, but studies done to date based upon visual inspections. In this study, an algorithm has been developed to help in evaluation of the disease. By the algorithm, brain is automatically detected, and all minimal distances of tissue pixels from dura are computed. According to normalized distances, average densities of certain sites are proportioned and a numerical criterion is calculated. This criterion is verified with MR images of three patients.