Automated Optic Nerve Sheath Diameter Measurement Using Super-pixel Analysis.

ABSTRACT

The optic nerve is a part of the central nervous system surrounded by cerebrospinal fluid and is encased in a sheath. Changes to the cerebrospinal fluid due to injury, tumor rupture and so on can increase intracranial pressure (ICP) and can result in changes in the sheath diameter. Measuring the changes in the sheath can be done through ultrasound imaging with which the optic nerve sheath diameter can be measured. Since this approach is non-invasive, it would reduce the cost for patients and healthcare if sheath diameter could be used as a predictor of increase in ICP. However, the manual measurement of the nerve sheath diameter is very time consuming and could be affected by human errors. In this paper we propose an image processing approach in which the optic nerve sheath diameter is measured automatically. In our proposed method, we first denoise images and then detect the region of interest using a simple line integral method. After that by analyzing super-pixels we measure the diameter. We compared the results of the proposed method with manual measurements from two experts. The average percentage of error between the proposed method and the experts' measurements did not substantially differ from the error between the two experts.