An efficient monitoring of eclamptic seizures in wireless sensors networks

Daniyal Haider, Aifeng Ren, Dou Fan, Nan Zhao, Xiaodong Yang, Syed Shah, Fangming Hu, Qammer H. Abbasi

Research output: Contribution to journalArticlepeer-review

30 Citations (Scopus)

Abstract

This paper presents the application of wireless sensing at C-band operating at 4.8 GHz technology (a potential Chinese 5G band). A wireless transceiver is used in the indoor environment to monitor different body motions of a woman experiencing an eclamptic seizure. The body movement shows unique wireless data which carries the wireless channel information. The results indicate that using higher features increases the accuracy from 3% to 4% for classifying data from different body motions. All of the four classifiers are compared by using six performance metrics such as accuracy, recall, precession, specificity, F-measure and Kappa. The values from these metrics indicate the better performance of SVM as compared to other three classifiers, the results indicate that the eclamptic seizures are easily differentiated from other body movements by applying the aforementioned classifiers.
Original languageEnglish
Pages (from-to)16-30
Number of pages15
JournalComputers and Electrical Engineering
Volume75
Early online date20 Feb 2019
DOIs
Publication statusPublished - 1 May 2019
Externally publishedYes

Keywords

  • C-band
  • Wireless channel information (WCI)
  • Internet of thing (IoT)
  • Support vector machine (SVM)
  • K-nearest neighbor (KNN)
  • Random forest (RF) and K-mean

Fingerprint

Dive into the research topics of 'An efficient monitoring of eclamptic seizures in wireless sensors networks'. Together they form a unique fingerprint.

Cite this