Zhilin Zhang
I am a staff research engineer and manager with the Emerging Technology Lab in Samsung Research America - Dallas, Samsung Electronics, as the main signal processing researcher for projects on wearable devices, smart-home, and health monitoring.
I received the Ph.D. degree in Electrical Engineering (Signal and Image Processing) from University of California, San Diego (UCSD) in 2012, under the supervision of Prof. Bhaskar D. Rao. More information about me is available at here.
My research interest includes Signal Processing, Data Analysis, and Machine Learning, particularly (details can be found at here):
(1) Statistical Signal Processing (weak signal detection and estimation)
(2) Sparse Signal Recovery & Compressed Sensing
(3) Biomedical Signal Processing & Neuroimaging (ECG, EEG, PPG, PCG, EMG, EOG, MRI)
(4) Machine Learning and Pattern Recognition
(5) Signal Separation and Decomposition
(6) Sensor Data Fusion (inertial sensors and biosensors)
(7) Financial Data Analysis and Forecasting
with their applications to
(a) Smart Home (smart device control, energy/power)
(b) Healthcare and Fitness (wearable devices development, mobile-based health monitoring)
Supervisors: Bhaskar D. Rao
I received the Ph.D. degree in Electrical Engineering (Signal and Image Processing) from University of California, San Diego (UCSD) in 2012, under the supervision of Prof. Bhaskar D. Rao. More information about me is available at here.
My research interest includes Signal Processing, Data Analysis, and Machine Learning, particularly (details can be found at here):
(1) Statistical Signal Processing (weak signal detection and estimation)
(2) Sparse Signal Recovery & Compressed Sensing
(3) Biomedical Signal Processing & Neuroimaging (ECG, EEG, PPG, PCG, EMG, EOG, MRI)
(4) Machine Learning and Pattern Recognition
(5) Signal Separation and Decomposition
(6) Sensor Data Fusion (inertial sensors and biosensors)
(7) Financial Data Analysis and Forecasting
with their applications to
(a) Smart Home (smart device control, energy/power)
(b) Healthcare and Fitness (wearable devices development, mobile-based health monitoring)
Supervisors: Bhaskar D. Rao
less
Related Authors
InterestsView All (8)
Uploads
Papers by Zhilin Zhang
(PPG) signals during subjects’ intensive exercise is
a difficult problem, since the PPG signals are contaminated
by extremely strong motion artifacts caused by subjects’ hand
movements. In this work, we formulate the heart rate estimation problem as a sparse signal recovery problem, and use a sparse
signal recovery algorithm to calculate high-resolution power
spectra of PPG signals, from which heart rates are estimated
by selecting corresponding spectrum peaks. To facilitate the use
of sparse signal recovery, we propose using bandpass filtering,
singular spectrum analysis, and temporal difference operation
to partially remove motion artifacts and sparsify PPG spectra.
The proposed method was tested on PPG recordings from 10
subjects who were fast running at the peak speed of 15km/hour.
The results showed that the averaged absolute estimation error
was only 2.56 Beats/Minute, or 1.94% error compared to groundtruth heart rates from simultaneously recorded ECG.
The TROIKA framework has high estimation accuracy and is robust to strong motion artifacts. Many variants can be straightforwardly derived from this framework. Experimental results on datasets recorded from 12 subjects during fast running at the peak speed of 15 km/hour showed that the average absolute error of heart rate estimation was 2.34 beat per minute (BPM), and the Pearson correlation between the estimates and the ground-truth of heart rate was 0.992. This framework is of great values to wearable devices such as smart-watches which use PPG signals to monitor heart rate for fitness.
This work proposes to use the block sparse Bayesian learning (BSBL) framework to compress/reconstruct non-sparse raw FECG recordings. Experimental results show that the framework can reconstruct the raw recordings with high quality. Especially, the reconstruction does not destroy the interdependence relation among the multichannel recordings. This ensures that the independent component analysis decomposition of the reconstructed recordings has high fidelity. Furthermore, the framework allows the use of a sparse binary sensing matrix with much fewer nonzero entries to compress recordings. Particularly, each column of the matrix can contain only two nonzero entries. This shows the framework, compared to other algorithms such as current CS algorithms and wavelet algorithms, can greatly reduce code execution in CPU in the data compression stage."
sensing has drawn much attention in wireless telemonitoring
of biosignals due to its ability to reduce energy consumption
and make possible the design of low-power devices. However,
the non-sparseness of biosignals presents a major challenge
to compressed sensing. This study proposes and evaluates a
spatio-temporal sparse Bayesian learning algorithm, which has
the desired ability to recover such non-sparse biosignals. It
exploits both temporal correlation in each individual biosignal
and inter-channel correlation among biosignals from different
channels. The proposed algorithm was used for compressed
sensing of multichannel electroencephalographic (EEG) signals
for estimating vehicle drivers’ drowsiness. Results showed that
the drowsiness estimation was almost unaffected even if raw
EEG signals (containing various artifacts) were compressed by
90%.
quantitative electroencephalography (EEG) distinguish among Alzheimer's Disease (AD) patients, mild cognitive impaired (MCI) subjects and elderly healthy controls? In other words, are there nonlinear indexes extracted from raw EEG data that are able to manifest the background difference among EEG? The response we give here is that a synthetic index of entropic complexity (Permutation Entropy, PE) as well as a measure of compressibility of the EEG can be used to discriminate among classes of subjects. An experimental database has been analyzed to make these measurements and the results we achieved are encouraging also in terms of disease evolution. Indeed, it is clearly shown that the condition of MCI has intermediate properties with respect to the analyzed markers: thus, these markers could in principle be used to evaluate the probability of transition from MCI to mild AD.
(PPG) signals during subjects’ intensive exercise is
a difficult problem, since the PPG signals are contaminated
by extremely strong motion artifacts caused by subjects’ hand
movements. In this work, we formulate the heart rate estimation problem as a sparse signal recovery problem, and use a sparse
signal recovery algorithm to calculate high-resolution power
spectra of PPG signals, from which heart rates are estimated
by selecting corresponding spectrum peaks. To facilitate the use
of sparse signal recovery, we propose using bandpass filtering,
singular spectrum analysis, and temporal difference operation
to partially remove motion artifacts and sparsify PPG spectra.
The proposed method was tested on PPG recordings from 10
subjects who were fast running at the peak speed of 15km/hour.
The results showed that the averaged absolute estimation error
was only 2.56 Beats/Minute, or 1.94% error compared to groundtruth heart rates from simultaneously recorded ECG.
The TROIKA framework has high estimation accuracy and is robust to strong motion artifacts. Many variants can be straightforwardly derived from this framework. Experimental results on datasets recorded from 12 subjects during fast running at the peak speed of 15 km/hour showed that the average absolute error of heart rate estimation was 2.34 beat per minute (BPM), and the Pearson correlation between the estimates and the ground-truth of heart rate was 0.992. This framework is of great values to wearable devices such as smart-watches which use PPG signals to monitor heart rate for fitness.
This work proposes to use the block sparse Bayesian learning (BSBL) framework to compress/reconstruct non-sparse raw FECG recordings. Experimental results show that the framework can reconstruct the raw recordings with high quality. Especially, the reconstruction does not destroy the interdependence relation among the multichannel recordings. This ensures that the independent component analysis decomposition of the reconstructed recordings has high fidelity. Furthermore, the framework allows the use of a sparse binary sensing matrix with much fewer nonzero entries to compress recordings. Particularly, each column of the matrix can contain only two nonzero entries. This shows the framework, compared to other algorithms such as current CS algorithms and wavelet algorithms, can greatly reduce code execution in CPU in the data compression stage."
sensing has drawn much attention in wireless telemonitoring
of biosignals due to its ability to reduce energy consumption
and make possible the design of low-power devices. However,
the non-sparseness of biosignals presents a major challenge
to compressed sensing. This study proposes and evaluates a
spatio-temporal sparse Bayesian learning algorithm, which has
the desired ability to recover such non-sparse biosignals. It
exploits both temporal correlation in each individual biosignal
and inter-channel correlation among biosignals from different
channels. The proposed algorithm was used for compressed
sensing of multichannel electroencephalographic (EEG) signals
for estimating vehicle drivers’ drowsiness. Results showed that
the drowsiness estimation was almost unaffected even if raw
EEG signals (containing various artifacts) were compressed by
90%.
quantitative electroencephalography (EEG) distinguish among Alzheimer's Disease (AD) patients, mild cognitive impaired (MCI) subjects and elderly healthy controls? In other words, are there nonlinear indexes extracted from raw EEG data that are able to manifest the background difference among EEG? The response we give here is that a synthetic index of entropic complexity (Permutation Entropy, PE) as well as a measure of compressibility of the EEG can be used to discriminate among classes of subjects. An experimental database has been analyzed to make these measurements and the results we achieved are encouraging also in terms of disease evolution. Indeed, it is clearly shown that the condition of MCI has intermediate properties with respect to the analyzed markers: thus, these markers could in principle be used to evaluate the probability of transition from MCI to mild AD.