视跟踪系统的信号处理

SIGNAL PROCESSING OF SMOOTH PURSUIT EYE-MOVEMENT SYSTEM

本文选择了眼动的视跟踪信号作为研究小脑机制的窗口，将临床记录的与小脑疾病有关的４０例病人作为对象（另加４０例正常人作为对照），根据眼动系统神经生理知识，提出信号处理的特征提取方法。将视跟踪信号分割为近似平稳跟踪信号（简称基干波）及近似扫视信号，将近似扫视信号部分的最大扫视速度、各扫视波扫视速度在１５～５００／ｓ、５０～８００／ｓ及大于８００／ｓ三个区段内的频数作为信号特征，在统计处理的基础上，发现几种小脑病人与正常人特征相比有显著性差异，揭示了小脑在视跟踪中的重要作用，从而为临床特别是功能性疾病的早期而又准确的诊断提供参考。

In this paper we choose eye tracking signal as approach to investigate the mechanism of cerebellum diaeases.Records of forty patients with cerebellar disorders were studied.Same numbers of normal person's records were used as control.The method to pick up signal features was basde on the knowledge in the neurophysiology of oculomotor system.The eye tracking signal was devided into approximate smooth pursuit (called the backbone wave for short)and approximate saccade.Maximal saccadic velocity,frequencies of occurrence in saccadic velocity in the range of 15-50、50-80 and above 80 degree per second were utilizded to specified the signal features.Statistically,we have found that the features of different types of cerebellar disorders were strikingly different from the normal,pointing out the cerebellum playing an important role in tracking eye movement.Having confirmed clinically,these findings may provide a clue for early diagnosis and assessments of functional disorders.