سامانه همایش‌ها TSTA | ثبت‌نام و ارسال مقاله کنفرانس‌ها و کنگره‌های ملی

Home / سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران

طراحی یک سیستم تشخیص سطح لرزش برای بیماران پارکینسون بر اساس توپولوژی سری زمانی لرزش در فضای فاز جغرافیایی

Authors :

مهدی ذوالفقارزاده کرمانی¹ سعید رشیدی² اساسه مریم³

1- گروه مهندسی پزشکی، واحد علوم و تحقیقات، دانشگاه آزاد اسلامی 2- گروه مهندسی پزشکی، واحد علوم و تحقیقات، دانشگاه آزاد اسلامی 3- گروه روانشناسی و آموزش کودکان استثنایی، واحد علوم و تحقیقات، دانشگاه آزاد اسلامی

Keywords :

Compass direction،Geographical phase space،Cartographical pattern،Tremor severity

Abstract :

We believe that the geometric structures or topologies of points reconstructed from time series in phase space contain richer information for classifying nonlinear time series; However, there is a lack of analytical tools to extract such information. In this study, a novel processing space based on geographical directions is presented, utilizing reconstructed trajectories from time series in phase space. Therefore, this geographic phase space provides a platform for extracting cartographical patterns related to different problems. This space was utilized in designing a tremor detection system for Parkinson's disease that classifies tremor signals into two levels: high-amplitude tremor and low-amplitude tremor. The findings of this study showed that a pattern consisting of north, east, and northwest directions reconstructed from the tremor signal was directly related to the severity of Parkinson's disease. Using the k-nearest neighbors classifier, the metrics of accuracy, sensitivity, and positive predictive value were calculated for training data equal to 100% while these metrics were obtained for the test data with accuracy = 96.55%, sensitivity = 98.25%, and positive predictive value = 95.44%. While the application of this new space is promising for identifying abnormal geographical directions associated with tremor severity in Parkinson's disease scenarios, it is recommended that future research examine the computational efficiency of the proposed space in processing other biomedical signals.