A5073700240- Neuroscience and Neural Engineering
- Wireless Power Transfer Systems
- Advanced Memory and Neural Computing
- Advanced Sensor and Energy Harvesting Materials
- Logic, programming, and type systems
- Perovskite Materials and Applications
- Markov Chains and Monte Carlo Methods
- EEG and Brain-Computer Interfaces
- Bayesian Methods and Mixture Models
- Energy Harvesting in Wireless Networks
- Conducting polymers and applications
- Formal Methods in Verification
- Microfluidic and Capillary Electrophoresis Applications
- Logic, Reasoning, and Knowledge
- Microfluidic and Bio-sensing Technologies
- Probabilistic and Robust Engineering Design
- Block Copolymer Self-Assembly
- Gaussian Processes and Bayesian Inference
- Domain Adaptation and Few-Shot Learning
- Virtual Reality Applications and Impacts
- Seismic Waves and Analysis
- Wireless Body Area Networks
- Interactive and Immersive Displays
- Visual Attention and Saliency Detection
- Augmented Reality Applications
Rice University
2018-2025
The University of Texas at Austin
2024
Nvidia (United States)
2023
Berkeley Geochronology Center
2020
University of St. Thomas - Texas
2019
Baylor College of Medicine
2019
Sensors (United States)
2017
University of California, Berkeley
2017
University of Canterbury
2016
Abstract Implantable bioelectronic devices for the simulation of peripheral nerves could be used to treat disorders that are resistant traditional pharmacological therapies. However, many nerve targets, this requires invasive surgeries and implantation bulky (about a few centimetres in at least one dimension). Here we report design vivo proof-of-concept testing an endovascular wireless battery-free millimetric implant stimulation specific difficult reach via surgeries. The device can...
Objective.Compared to biomedical devices with implanted batteries, wirelessly powered technologies can be longer-lasting, less invasive, safer, and miniaturized access difficult-to-reach areas of the body. Magnetic fields are an attractive wireless power transfer modality for such bioelectronic applications because they suffer negligible absorption reflection in biological tissues. However, current solutions using magnetic mm sized implants either operate at high frequencies (>500 kHz) or...
Abstract Microfluidic concentration gradient generators ( µ -CGGs) have been utilized to identify optimal drug compositions through antimicrobial susceptibility testing (AST) for the treatment of antimicrobial-resistant (AMR) infections. Conventional -CGGs fabricated via photolithography-based micromachining processes, however, are fundamentally limited two-dimensional fluidic routing, such that only two distinct drugs can be tested at once. This work addresses this limitation by employing...
This article presents a hardware platform including stimulating implants wirelessly powered and controlled by shared transmitter (TX) for coordinated leadless multisite stimulation. The adopted novel single-TX, multiple-implant structure can flexibly deploy stimuli, improve system efficiency, easily scale channel quantity, relieve efforts in device synchronization. In the proposed system, wireless link leveraging magnetoelectric (ME) effect is co-designed with robust efficient system-on-chip...
Designing implantable bioelectronic systems that continuously monitor physiological functions and simultaneously provide personalized therapeutic solutions for patients remains a persistent challenge across many applications ranging from neural to organs. Closed-loop typically consist of three functional blocks, namely, sensors, signal processors actuators. An effective system, can the necessary therapeutics, tailored individual factors requires distributed network sensors While significant...
Modulating the electrical activity in nervous system has shown great potential for neuroscience research and clinical therapies. To reduce risks of infection restrictions subject mobility, neuromodulators must be miniaturized untethered. Safe reliable wireless power transfer data delivery with required size constraints is still one fundamental challenges developing miniature neural interfaces. A few implants powered by RF, inductive coupling [1]-[3], ultrasound [4], optics [5] have been...
This paper presents the first wireless and programmable neural stimulator leveraging magnetoelectric (ME) effects for power data transfer. Thanks to low tissue absorption, misalignment sensitivity high transfer efficiency, ME effect enables safe delivery of levels (a few milliwatts) at resonant frequencies (~250 kHz) mm-sized implants deep inside body (30-mm depth). The presented MagNI (Magnetoelectric Neural Implant) consists a 1.5-mm$^2$ 180-nm CMOS chip, an in-house built 4x2 mm film,...
This paper presents a symmetric three-flow microfluidic concentration gradient generator, made possible by 3D microchannel network fabricated via ultra-high resolution Multijet printing. The prototype fluidic device is utilized as an effective tool for the expedited screening of multi-drug combinations. It rapidly generates 15 discrete combinations three input drugs which are used to determine minimum inhibitory (MIC) value individual antibiotics, well perform simultaneous clinical three-way...
We propose a fast and scalable variational method for Bayesian inference in high-dimensional parameter space, which we call projected Stein Newton (pSVN) method. exploit the intrinsic low-dimensional geometric structure of posterior distribution space via its Hessian (of log posterior) operator perform parallel update samples into subspace by an SVN The is adaptively constructed using eigenvectors averaged at current samples. demonstrate convergence proposed scalability with respect to...
Summary form only given. Implantable bioelectronics for electrically modulating activities of specific cells have shown great success and exciting potential in treating a wide range diseases. Some the most representative therapies are cardiac pacemakers neuromodulators motor function restoration, pain relief neural disorder treatment. While several wireless miniaturized bio-stimulators been demonstrated, them lack capability coordinated multisite stimulation, which is to be more effective...
This paper presents a hardware platform for wireless mm-sized bio-implant networks, exploiting adaptive magnetoelectric power transfer and novel schemes efficient bidirectional multi-access communication. The closed-loop control mitigates delivery fluctuations caused by distance alignment change avoids redundant of the external transceiver. system also enables simultaneous time-domain modulated downlink data with 5% peak efficiency 62.3-kbps maximum rate at 340-kHz carrier frequency; uplink...
Stein variational gradient descent (SVGD) is a general-purpose optimization-based sampling algorithm that has recently exploded in popularity, but limited by two issues: it known to produce biased samples, and can be slow converge on complicated distributions. A proposed stochastic variant of SVGD (sSVGD) addresses the first issue, producing unbiased samples incorporating special noise into dynamics such asymptotic convergence guaranteed. Meanwhile, Newton (SVN), Newton-like extension SVGD,...
Watching spherical panorama movies is one of the applications head mounted displays, especially growing number lowcost consumer devices. In this poster, we show how to enhance a personal immersive cinematic experience by embedding user's body in movie scene using Augmented Virtuality technology. User embodiment, transitioning between real and virtual spaces, adding interactivity are main benefits our approach. We present proof concept prototype, summarize findings from focus group held...
Abstract A fundamental challenge for bioelectronics is to deliver power miniature devices inside the body. Wires are common failure points and limit device placement. On other hand, wireless by electromagnetic or ultrasound waves must overcome absorption body impedance mismatches between air, bone, tissue. In contrast, magnetic fields suffer little differences in at interfaces These advantages have led magnetically-powered stimulators based on induction magnetothermal effects. However,...
We consider the Bayesian calibration of models describing phenomenon block copolymer (BCP) self-assembly using image data produced by microscopy or X-ray scattering techniques. To account for random long-range disorder in BCP equilibrium structures, we introduce auxiliary variables to represent this aleatory uncertainty. These variables, however, result an integrated likelihood high-dimensional that is generally intractable evaluate. tackle challenging inference problem a likelihood-free...
3D Object detection is a fundamental task in vision-based autonomous driving. Deep learning perception models achieve an outstanding performance at the expense of continuously increasing resource needs and, as such, training costs. As inference time still priority, developers usually adopt pipeline where they first start using compact architecture that yields good trade-off between accuracy and latency. This found either by searching manually or neural search approaches. Then, train model...
Abstract Minimally invasive neural interfaces can be used to diagnose, manage, and treat many disorders with substantially reduced risks of surgical complications. Endovascular implanted in the veins or arteries is one approach, but it requires prescriptions anti-thrombotic medication are likely not explantable after endothelialization. More critically, approach limited by small size location blood vessels, such that important cortical, subcortical, spinal targets cannot reached. Here, we...