STN area detection using K-NN classifiers for MER recordings in Parkinson patients during neurostimulator implant surgery

20th Argentinean Bioengineering Society Congress

物理学;生物科学

Schiaffino, L.^1 ; Rosado Munoz, A.^2 ; Guerrero Martinez, J.^2 ; Francés Villora, J.^2 ; Gutiérrez, A.^3 ; Martinez Torres, I.^3 ; Kohan, Y.D.R.^4

Rehabilitation Engineering and Neuromuscular and Sensorial Research Laboratory (LIRINS), National University of Entre Rios, Argentina^1

Digital Signal Processing Group, ETSE, Department of Electronic Engineering, University of Valencia, Valencia, Spain^2

Functional Neurosurgery Unit. la Fe Hospital, Valencia, Spain^3

Department of Probability and Statistics, Faculty of Engineering, National University of Entre Rios, Argentina^4

关键词: Deep brain stimulation;    Detection algorithm;    K nearest neighbours (k-NN);    Microelectrode recording;    Parkinson's disease;    Sensitivity and specificity;    Statistical hypothesis;    Stimulating electrodes;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/705/1/012050/pdf
DOI  :  10.1088/1742-6596/705/1/012050

学科分类：生物科学（综合）

来源: IOP

【 摘 要 】

Deep Brain Stimulation (DBS) applies electric pulses into the subthalamic nucleus (STN) improving tremor and other symptoms associated to Parkinson's disease. Accurate STN detection for proper location and implant of the stimulating electrodes is a complex task and surgeons are not always certain about final location. Signals from the STN acquired during DBS surgery are obtained with microelectrodes, having specific characteristics differing from other brain areas. Using supervised learning, a trained model based on previous microelectrode recordings (MER) can be obtained, being able to successfully classify the STN area for new MER signals. The K Nearest Neighbours (K-NN) algorithm has been successfully applied to STN detection. However, the use of the fuzzy form of the K-NN algorithm (KNN-F) has not been reported. This work compares the STN detection algorithm of K-NN and KNN-F. Real MER recordings from eight patients where previously classified by neurophysiologists, defining 15 features. Sensitivity and specificity for the classifiers are obtained, Wilcoxon signed rank non-parametric test is used as statistical hypothesis validation. We conclude that the performance of KNN-F classifier is higher than K-NN with p