A5100639668- Neuroscience and Neural Engineering
- EEG and Brain-Computer Interfaces
- Advanced Memory and Neural Computing
- Muscle activation and electromyography studies
- Neural dynamics and brain function
- Chaos-based Image/Signal Encryption
- Photoreceptor and optogenetics research
- Wireless Power Transfer Systems
- Analog and Mixed-Signal Circuit Design
- Advanced Sensor and Energy Harvesting Materials
- Advanced Steganography and Watermarking Techniques
- Conducting polymers and applications
- Analytical Chemistry and Sensors
- CCD and CMOS Imaging Sensors
- Sleep and Wakefulness Research
- Blind Source Separation Techniques
- Energy Harvesting in Wireless Networks
- Wireless Body Area Networks
- Cellular Automata and Applications
- Advanced MIMO Systems Optimization
- Quantum-Dot Cellular Automata
- Hydraulic Fracturing and Reservoir Analysis
- Context-Aware Activity Recognition Systems
- Neurological disorders and treatments
- Millimeter-Wave Propagation and Modeling
University of Toronto
2021-2025
University Health Network
2023-2024
Sichuan Academy of Medical Sciences & Sichuan Provincial People's Hospital
2022-2024
Toronto Rehabilitation Institute
2023-2024
Harbin Institute of Technology
2010-2024
First Affiliated Hospital of Wannan Medical College
2022
Wannan Medical College
2022
Second Military Medical University
2022
Lanzhou Jiaotong University
2022
University of Pennsylvania
2013-2021
Reliable, multi-channel neural recording is critical to the neuroscience research and clinical treatment. However, most hardware development of fully integrated, wireless recorders to-date, still in proof-of-concept stage. To be ready for practical use, trade-offs between performance, power consumption, device size, robustness, compatibility need carefully taken into account. This paper presents an optimized compressed sensing signal system. The system takes advantages both custom integrated...
This paper presents a bidirectional brain machine interface (BMI) microsystem designed for closed-loop neuroscience research, especially experiments in freely behaving animals. The system-on-chip (SoC) consists of 16-channel neural recording front-ends, feature extraction units, programmable stimulator back-ends, in-channel controllers, global analog-digital converters (ADC), and peripheral circuits. proposed units includes 1) an ultra low-power energy unit enabling 64-step natural...
This brief presents a wireless, low-power embedded system that recognizes hand gestures by decoding surface electromyography (EMG) signals. Ten used on commercial trackpads, including pinch, stretch, swipe left, right, scroll up, down, single click, double pat, and ok, can be recognized in real time. Features from four differential EMG channels are extracted multiple time windows. Unlike traditional data segmentation methods, an event-driven method is proposed, with the gesture event...
Freezing of gait (FoG) is a debilitating symptom Parkinson's disease (PD). This work develops flexible wearable sensors that can detect FoG and alert patients companions to help prevent falls. detected on the using deep learning (DL) model with multi-modal sensory inputs collected from distributed wireless sensors. Two types are developed, including: 1) C-shape central node placed around patient's ears, which collects electroencephalogram (EEG), detects an on-device DL model, generates...
In this paper, a general purpose wireless Brain- Machine -Brain Interface (BMBI) system is presented. The integrates four battery-powered devices for the implementation of closed-loop sensorimotor neural interface, including signal analyzer, stimulator, body-area sensor node and graphic user interface implemented on PC end. analyzer features channel analog front-end with configurable bandpass filter, gain stage, digitization resolution, sampling rate. target frequency band from EEG to single...
Implanted devices providing real-time neural activity classification and control are increasingly used to treat neurological disorders, such as epilepsy Parkinson's disease. Classification performance is critical identifying brain states appropriate for the therapeutic action. However, advanced algorithms that have shown promise in offline studies, particular deep learning (DL) methods, not been deployed on resource-restrained implants. Here, we designed optimized three embedded DL models of...
The peripheral nervous system (PNS) provides a conduit through which organs can communicate with the central system. PNS neural interfaces have been deployed in open-loop fashion to help restore motor or sensory functions paralyzed amputated individuals, and also as implantable closed-loop therapeutic devices for treating chronic medical conditions related autoimmune metabolic disorders. Their efficacy scope of clinical use, however, are severely curtailed by invasiveness cable, electronics...
This paper presents a lightweight deep learning (DL) model for classifying sleep stages based on single-channel EEG. The DL was designed to run energy- and memory-constrained devices real-time operation with all processing the edge. Four convolutional filter layers are used extract features reduce data dimension, transformers were utilized learn time-variant of data. EEG recordings from publicly available dataset (Sleep-EDF) train test model. Subject-specific training implemented improve...
Recognizing patterns in lung sounds is crucial to detecting and monitoring respiratory diseases. Current techniques for analyzing demand domain experts are subject interpretation. Hence an accurate automatic sound classification system desired. In this work, we took a data-driven approach classify abnormal sounds. We compared the performance using three different feature extraction techniques, which short-time Fourier transformation (STFT), Mel spectrograms, Wav2vec, as well classifiers,...
In this paper, a general purpose wireless Brain-Machine-Brain Interface (BMBI) system is proposed. The provides all the necessary hardware for closed-loop sensorimotor neural interface. integrates signal analyzer, two stimulators with different specifications, multiple body area sensory devices and user-friendly computer analyzer features four channel analog frontend configurable bandpass filter, gain stage, digitization resolution, sampling rate. Digital filtering, feature extraction, spike...
Neuron apoptosis is a feature of secondary injury after traumatic brain (TBI). Evidence implies that excess calcium (Ca2+) ions and reactive oxidative species (ROS) play critical roles in apoptosis. In reaction to increased ROS, the anti-oxidative master transcription factor, Transient receptor potential Ankyrin 1 (TRPA1) allows Ca2+ enter cells. However, effect TBI on expression TRPA1 role are unclear. present study, mouse was simulated using weight-drop model. The process neuronal stress...
This article presents a wireless sensor-brain interface (SBI) system designed for closed-loop neuromodulation in freely behaving animals. The enables novel experiment which swimming rats navigate water maze guided exclusively by neural stimulation. consists of two wirelessly linked application-specific integrated circuits (ASICs): bidirectional and an animal-tracking image sensor. ASIC features stimulator design with adaptive termination-based charge-balancing energy-efficient recording...
This paper presents a 12-channel, low-power, high efficiency neural signal acquisition front-end for local field potential and action signals recording. The proposed integrates low noise instrumentation amplifiers, low-power filter stages with configurable gain cut-off frequencies, successive approximation register (SAR) ADC, realtime compressed sensing processing unit. A capacitor coupled amplifier integrated input impedance boosting has been designed, dissipating 1μA quiescent current. An...
Sensory feedback is critical to the performance of neural prostheses that restore movement control after neurological injury. Recent advances in direct paralyzed arms present new requirements for miniaturized, low-power sensor systems. To address this challenge, we developed a fully-integrated wireless sensor-brain-machine interface (SBMI) system communicating key somatosensory signals, fingertip forces and limb joint angles, brain. The consists tactile force sensor, an electrogoniometer,...
This paper presents the design of a wireless sensor network for detecting and alerting freezing gait (FoG) symptoms in patients with Parkinson's disease. A novel button pin type node was developed easy attachment. Three nodes, each integrating 3-axis accelerometer, can be placed on patient at their ankle, thigh, truck. Each independently detect FoG using an on-device deep learning (DL) model, featuring convolutional neural (CNN). The DL model outputs from three nodes are processed central...
Significance Our sensory experience is governed by sensor properties (e.g., eye photoreceptors) and corresponding motor strategies to sample the environment movements). With injury, aging, or new task constraints, existing can become incompatible with perceptual demands. Using a brain–machine interface paradigm in rats, we studied how are adapted inputs accomplish difficult searching task. We show that be dynamically regulated through optimally extract task-relevant information.