Shuang Sun

Shuang Sun

Auto-generated profile, based on publicly available information

Shuang Sun

College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 June 2021 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
Junxin ChenJunxin Chen
Junxin Chen
SSShuang Sun
Shuang Sun
LZLi-Bo Zhang
Li-Bo Zhang
BYBenqiang Yang
Benqiang Yang
WWWei Wang
Wei Wang
For continuous monitoring of cardiovascular diseases, this article presents a novel framework for heart sound acquisition. The proposed approach uses compressed sensing for signal sampling, and a two-stage reconstruction is developed for reconstruction. The first stage aims to give a tentative recovered signal, on which a peak detection technique is developed to identify whether there is a peak in current segment and, if so, its location. With such information, an adaptive dictionary is selected for the second round reconstruction. Because the selected dictionary is adaptive to the morphology of current frame, the signal reconstruction performance is consequently promoted. Experiment results indicate that a satisfactory performance can be obtained when the frame length is 256 and the signal morphology is divided into 16 categories. Furthermore, the proposed algorithm is compared with a series of counterparts, and the results well demonstrate the advantages of our proposal, especially at high compression ratios.

29 January 2021 in IEEE Sensors Journal
Verified
Journal Article
Closed access
Junxin ChenJunxin Chen
Junxin Chen
SSShuang Sun
Shuang Sun
NBNan Bao
Nan Bao
ZZZhi-liang Zhu
Zhi-liang Zhu
LZLi-Bo Zhang
Li-Bo Zhang
This paper presents an improved reconstruction approach for compressed sensing based ECG acquisition in Internet of Medical Things. The proposed method exploits the concepts of adaptive overcomplete dictionary and QRS detection in CS domain. Based on whether there is a QRS complex, the ECG frames to be reconstructed are divided into several categories and corresponding overcomplete dictionaries are trained to fit these different kinds of ECG frames. Specifically, QRS detection is first performed directly on the compressed measurements to determine the QRS morphology without actually reconstructing the signal in advance, and then a suitable overcomplete dictionary is chosen for the signal reconstruction. Because the selected dictionaries well fit the ECG frames, the reconstruction quality is consequently improved. Comparative experiments have been conducted, and the results well demonstrate the performance of the proposed approach.