Nystagmus Signal Feature Extraction And Its Application In The Vestibular System

Vestibular system plays an important role in the perception of human position. The vestibular can detect body position changes, pass this information to the eyes, and make the eyes move to direction which is opposite to the rotation direction of head, in order to maintain the stability of retinal imaging. But the vestibular may also pass the wrong information when vestibule function is abnormal, and make the eyes move to the wrong direction. This disorder makes people feel vertigo. For the reason that the vestibular is located in the inner ear, it is too deep and small to be observed directly. Vestibular detections have to use the special inspection equipments. But these inspection equipments have limitations due to the high cost and unfitting for children. Eye movement system and vestibular system has close coupling relationship. Although the medical professions have do a lot of research on the interaction of eye movement and vestibular function system, there are not effective methods to detection vestibular disease based on nystagmus signal feature. The research is lack of the other subjects, especially information subject, so it cannot give the nystagmus signals’quantitative mathematical description as so far. In this paper, according to the pathology analysis, we extract the features of nystagmus signal, such as torsional angle, slow phase, nystagmus direction, using signal processing technology, these features are applied to the diagnosis of vestibular diseases and its function study.In this paper, the main work includes:1) Using the image processing method and the pupil tracking algorithm to extract the nystagmus signal. Experimental results show that the algorithm can accurately locate the pupil center, which can meet the demand of clinical.2) Extracting the new feature of torsional nystagmus, torsional angle. Since the direction of torsional angle can be detected easily, it can be used as an objective auxiliary parameter for clinical diagnosis of torsional nystagmus. Another advantage of this algorithm is its low computational complexity.3) An automatic diagnosis method for benign paroxysmal positional vertigo (BPPV) based on the new feature, torsional angle is proposed. The diagnosis results show that the proposed method is an efficient and accurate method for clinical diagnosis of BPPV. 4) Establishing the model parameter of healthy person’s vestibular system, and applying this parameter to distinguish the type and degree of motion sickness. The model parameter of healthy person’s vestibular system is identified by rotating chair test. Study finds that for different types of subjects, the model parameter has obvious difference, especially between people with motion sickness and normal people.