KIUP KIM

SEUNGHYEON NAM

SANGJOON PARK

HYUNGKOOK KIM

JIWON SEO

INUG YOON

SEONGYEON KIM

GAYOUNG EOM

MUHAMMAD SARMAD

CHAEYUN SHIM

BYUNGBUM KANG

SUNGWOO LEE

MIKYUNG KIM

HYOJUNG KIM

SOL HAN

WONJUN LEE

The Brain/Bio Medical Microsystems Lab at KAIST focuses on neural interfaces, ultrasound neuromodulation, and biomedical sensors.

Flexible ultrasound transducer array with statically adjustable curvature for anti-inflammatory treatment

Low-intensity focused ultrasound enables temporal modulation of human midbrain organoid differentiation

Ultrasound brain stimulation technologies for targeted therapeutics

Cu/dielectric hybrid bonding: Experimental and simulation perspectives

Selective manipulation of excitatory and inhibitory neurons in top-down and bottom-up visual pathways using ultrasound stimulation

Fabrication of Capacitive Micromachined Ultrasonic Transducers With High-k Insulation Layer Using Silicon Fusion Bonding

Turning integrated circuits into ultrasound transducers

Highly stretchable 3D microelectrode array for non-invasive functional evaluation of cardiac spheroids and midbrain organoids

Microfabricated sensors for non-invasive, real-time monitoring of organoids

The Role of Artificial Intelligence in Obesity Medicine

Transcranial focused ultrasound stimulation in the infralimbic cortex facilitates extinction of conditioned fear in rats

An Ultrasound Receiver With Bandwidth-Enhanced Current Conveyor and Element-Level Ultrasound Transmitter for Ultrasound Imaging Systems

Self-Charging Dual-Modal Sensor for Glucose Monitoring Based on Piezoelectric Nanowire/Mircogel Hybrid Film

A Glucose-responsive Microgel-based Soft Etalon as an Epidermal Glucose Colorimetric Sensor

Miniaturized MR-Compatible Ultrasound System for Real-Time Monitoring of Acoustic Effects in Mice using High-Resolution MRI

Optimization and Structural Modeling of Pre-snapback Mode CMUT for Enhanced Acoustic Performance (Poster presentation)

Single-bias Dual-mode CMUT Array Utilizing Pre-snapback Region for Acoustic Hologram (Oral presentation)

Conformal Ultrasound Patch for Real-Time Blood Flow Monitoring

Development of Flex-to-Rigid Capacitive Micromachined Ultrasound Transducer (CMUT) with Bending Modulation

Calcium-Modified Silk Adhesive as Coupling Medium for Next-Generation Ultrasound Patch Applications

CMUT Humidity Sensor with a Calcium-Modified Silk Functionalized Layer

A flexible neural probe with electrodeposited silk composite as an insertion shuttle

Calcium-Modified Silk FIlm as a Patch-Type Ultrasound Coupling Medium

An integrated circuit with transmit beamforming and parallel receive channels for real-time three-dimensional ultrasound imaging

An integrated circuit with transmit beamforming and parallel receive channels for 3D ultrasound imaging: Testing and characterization

Fabricating capacitive micromachined ultrasonic transducers with direct wafer-bonding and LOCOS technology

CMUT as a chemical sensor for DMMP Detection

Optimum design of circular CMUT membranes for high quality factor in air

A low-noise oscillator based on a multi-membrane CMUT for high sensitivity resonant chemical sensors

Modeling and measuring the effects of mutual impedance on multi-cell CMUT configurations

Achieving minimum-cost multicast: A decentralized approach based on network coding

RL-GAN-Net: A Reinforcement Learning Agent Controlled GAN Network for Real-Time Point Cloud Shape Completion

Calcium‐Modified Silk as a Biocompatible and Strong Adhesive for Epidermal Electronics

Safety of transcranial focused ultrasound stimulation: A systematic review of the state of knowledge from both human and animal studies

An integrated circuit with transmit beamforming flip-chip bonded to a 2-D CMUT array for 3-D ultrasound imaging

Multifunctional multi-shank neural probe for investigating and modulating long-range neural circuits in vivo

BRAF somatic mutation contributes to intrinsic epileptogenicity in pediatric brain tumors

Aberrant Tonic Inhibition of Dopaminergic Neuronal Activity Causes Motor Symptoms in Animal Models of Parkinson’s Disease

In vivo optical modulation of neural signals using monolithically integrated two-dimensional neural probe arrays

Fabrication of capacitive micromachined ultrasonic transducers via local oxidation and direct wafer bonding

MEMS devices for drug delivery

A multichannel neural probe with embedded microfluidic channels for simultaneous in vivo neural recording and drug delivery

Capacitive micromachined ultrasonic transducers for chemical detection in nitrogen

Capacitive micromachined ultrasonic transducer (CMUT) as a chemical sensor for DMMP detection

Transcranial focused ultrasound stimulation with high spatial resolution

Chemical vapor detection using a capacitive micromachined ultrasonic transducer

CMUT-based resonant gas sensor array for VOC detection with low operating voltage

Neural probes with multi-drug delivery capability

Artifact‐Free 2D Mapping of Neural Activity In Vivo through Transparent Gold Nanonetwork Array

Mesoporous thin-film on highly-sensitive resonant chemical sensor for relative humidity and CO2 detection

Il-Joo Cho

Mario Kupnik

Ömer Oralkan

Eui-Sung Yoon

Hans Peter Lang

Goksen Goksenin Yaralioglu

Arif Sanli Ergun

Hyunjoo Jenny Lee

We present decentralized algorithms that compute minimum-cost subgraphs for establishing multicast connections in networks that use coding. These algorithms, coupled with existing decentralized schemes for constructing network codes, constitute a fully decentralized approach for achieving minimum-cost multicast. Our approach is in sharp contrast to the prevailing approach based on approximation algorithms for the directed Steiner tree problem, which is suboptimal and generally assumes centralized computation with full network knowledge. We also give extensions beyond the basic problem of fixed-rate multicast in networks with directed point-to-point links, and consider the case of elastic rate demand as well as the problem of minimum-energy multicast in wireless networks.

We present RL-GAN-Net, where a reinforcement learning (RL) agent provides fast and robust control of a generative adversarial network (GAN). Our framework is applied to point cloud shape completion that converts noisy, partial point cloud data into a high-fidelity completed shape by controlling the GAN. While a GAN is unstable and hard to train, we circumvent the problem by (1) training the GAN on the latent space representation whose dimension is reduced compared to the raw point cloud input and (2) using an RL agent to find the correct input to the GAN to generate the latent space representation of the shape that best fits the current input of incomplete point cloud. The suggested pipeline robustly completes point cloud with large missing regions. To the best of our knowledge, this is the first attempt to train an RL agent to control the GAN, which effectively learns the highly nonlinear mapping from the input noise of the GAN to the latent space of point cloud. The RL agent replaces the need for complex optimization and consequently makes our technique real time. Additionally, we demonstrate that our pipelines can be used to enhance the classification accuracy of point cloud with missing data.

With the increasing interest and demand for epidermal electronics, a strong interface between a sensor and a biological surface is essential, yet achieving such interface is still a challenge. Here, a calcium (Ca)‐modified biocompatible silk fibroin as a strong adhesive for epidermal electronics is proposed and the physical principles behind its interfacial and adhesive properties are reported. A strong adhesive characteristic (>800 N m −1 ) is observed because of the increase in both viscoelastic property and mechanical interlocking through the incorporation of Ca ions. Furthermore, additional key characteristics of the Ca‐modified silk: reusability, stretchability, biocompatibility, and conductivity, are reported. These characteristics enable a wide range of applications as demonstrated in four epidermal electronic systems: capacitive touch sensor, resistive strain sensor, hydrogel‐based drug delivery, and …

Background Low-intensity transcranial focused ultrasound stimulation (TFUS) holds great promise as a highly focal technique for transcranial stimulation even for deep brain areas. Yet, knowledge about the safety of this novel technique is still limited. Objective To systematically review safety related aspects of TFUS. The review covers the mechanisms-of-action by which TFUS may cause adverse effects and the available data on the possible occurrence of such effects in animal and human studies. Methods Initial screening used key term searches in PubMed and bioRxiv, and a review of the literature lists of relevant papers. We included only studies where safety assessment was performed, and this results in 33 studies, both in humans and animals. Results Adverse effects of TFUS were very rare. At high stimulation intensity and/or rate, TFUS may cause haemorrhage, cell death or damage, and unintentional blood …

State-of-the-art 3-D medical ultrasound imaging requires transmitting and receiving ultrasound using a 2-D array of ultrasound transducers with hundreds or thousands of elements. A tight combination of the transducer array with integrated circuitry eliminates bulky cables connecting the elements of the transducer array to a separate system of electronics. Furthermore, preamplifiers located close to the array can lead to improved receive sensitivity. A combined IC and transducer array can lead to a portable, high-performance, and inexpensive 3-D ultrasound imaging system. This paper presents an IC flip-chip bonded to a 16times16-element capacitive micromachined ultrasonic transducer (CMUT) array for 3-D ultrasound imaging. The IC includes a transmit beamformer that generates 25-V unipolar pulses with programmable focusing delays to 224 of the 256 transducer elements. One-shot circuits allow adjustment …

Investigation and modulation of neural circuits in vivo at the cellular level are very important for studying functional connectivity in a brain. Recently, neural probes with stimulation capabilities have been introduced, and they provided an opportunity for studying neural activities at a specific region in the brain using various stimuli. However, previous methods have a limitation in dissecting long-range neural circuits due to inherent limitations on their designs. Moreover, the large size of the previously reported probes induces more significant tissue damage. Herein, we present a multifunctional multi-shank MEMS neural probe that is monolithically integrated with an optical waveguide for optical stimulation, microfluidic channels for drug delivery, and microelectrode arrays for recording neural signals from different regions at the cellular level. In this work, we successfully demonstrated the functionality of our probe by …

Pediatric brain tumors are highly associated with epileptic seizures. However, their epileptogenic mechanisms remain unclear. Here, we show that the oncogenic BRAF somatic mutation p.Val600Glu (V600E) in developing neurons underlies intrinsic epileptogenicity in ganglioglioma, one of the leading causes of intractable epilepsy. To do so, we developed a mouse model harboring the BRAF V600E somatic mutation during early brain development to reflect the most frequent mutation, as well as the origin and timing thereof. Therein, the BRAF V600E mutation arising in progenitor cells during brain development led to the acquisition of intrinsic epileptogenic properties in neuronal lineage cells, whereas tumorigenic properties were attributed to high proliferation of glial lineage cells. RNA sequencing analysis of patient brain tissues with the mutation revealed that BRAF V600E -induced epileptogenesis is mediated by …

Current pharmacological treatments for Parkinson's disease (PD) are focused on symptomatic relief, but not on disease modification, based on the strong belief that PD is caused by irreversible dopaminergic neuronal death. Thus, the concept of the presence of dormant dopaminergic neurons and its possibility as the disease-modifying therapeutic target against PD have not been explored. Here we show that optogenetic activation of substantia nigra pars compacta (SNpc) neurons alleviates parkinsonism in acute PD animal models by recovering tyrosine hydroxylase (TH) from the TH-negative dormant dopaminergic neurons, some of which still express DOPA decarboxylase (DDC). The TH loss depends on reduced dopaminergic neuronal firing under aberrant tonic inhibition, which is attributed to excessive astrocytic GABA. Blocking the astrocytic GABA synthesis recapitulates the therapeutic effect of optogenetic …

Integration of stimulation modalities ( e.g. electrical, optical and chemical) on a large array of neural probes can enable an investigation of important underlying mechanisms of brain disorders that is not possible through neural recordings alone. Furthermore, it is important to achieve this integration of multiple functionalities in a compact structure to utilize a large number of the mouse models. Here we present a successful optical modulation of in vivo neural signals of a transgenic mouse through our compact 2D MEMS neural array (optrodes). Using a novel fabrication method that embeds a lower cladding layer in a silicon substrate, we achieved a thin silicon 2D optrode array that is capable of delivering light to multiple sites using SU-8 as a waveguide core. Without additional modification to the microelectrodes, the measured impedance of the multiple microelectrodes was below 1 MΩ at 1 kHz. In addition, with a low …

We present the successful fabrication of capacitive micromachined ultrasonic transducers (CMUTs) with an improved insulation layer structure. The goal is to improve device reliability (electrical breakdown) and device performance (reduced parasitic capacitance). The fabrication is based on consecutive thermal oxidation steps, on local oxidation of silicon (LOCOS), and on direct wafer bonding. No chemical-mechanical polishing step is required during the device fabrication. Aside from the advantages associated with direct wafer bonding for CMUT fabrication (simple fabrication, cell shape flexibility, wide gap height range, good uniformity, well-known material properties of single-crystal materials, and low intrinsic stress), the main vertical dimension (electrode separation) is determined by thermal oxidation only, which provides excellent vertical tolerance control ( <;10 nm) and unprecedented uniformity across the …

Novel drug delivery systems based on microtechnology have advanced tremendously, but yet face some technological and societal hurdles to fully achieve their potential. The novel drug delivery systems aim to deliver drugs in a spatiotemporal- and dosage-controlled manner with a goal to address the unmet medical needs from oral delivery and hypodermic injection. The unmet needs include effective delivery of new types of drug candidates that are otherwise insoluble and unstable, targeted delivery to areas protected by barriers ( e . g . brain and posterior eye segment), localized delivery of potent drugs, and improved patient compliance. After scrutinizing the design considerations and challenges associated with delivery to areas that cannot be efficiently targeted through standard drug delivery ( e . g . brain, posterior eye segment, and gastrointestinal tract), this review provides a summary of recent advances that …

Multi-functional neural probes integrated with various stimulation modalities are becoming essential tools in neuroscience to study the brain more effectively. In this paper, we present a new multi-functional neural probe that allows chemical stimulation through drug delivery and simultaneous recording of individual neuron signals through a microelectrode array. By embedding microchannels in silicon using a proposed glass reflow process, we successfully fabricated 40 μm thick silicon neural probes suitable for small animal experiments. The electrochemical impedance spectroscopy confirms that impedance of iridium microelectrodes is low enough (<1 MΩ at 1 kHz) to measure neural signals. Flow rate characterization in a 0.9% w/v agarose gel shows the capability to deliver a small volume of drugs (<1 μl) at a controlled flow rate. We demonstrate the viability and potential of this new probe by conducting in vivo …

The authors present the prototype of a chemical sensor using a capacitive micromachined ultrasonic transducer array. Each element in the array consists of a large number of resonating membranes connected in parallel. A five-channel oscillator circuit operates at the resonant frequency around 6 MHz in this prototype. The surface of the elements in the array is coated by polymers such as polyallylamine hydrochloride, polyethylene glycol, and polyvinyl alcohol to detect different chemicals. By measuring shift in oscillation frequencies due to the mass-loading effect, analytes, eg, water and isopropanol, with concentrations around 20 ppbv (parts per 10 9 by volume) range can be detected.

We present a chemical sensor based on a capacitive micromachined ultrasonic transducer (CMUT) configured as a resonant mass sensor with a chemically selective polymer coating. The sensing unit of the CMUT consists of 100s to 1000s of resonators connected in parallel and acts as a single resonator. The high resonant frequency (18.2MHz) and the small mass (296pg) enable the CMUT to have a good mass sensitivity of 130zg/Hz/μm 2 . We functionalized the CMUT with polyisobutylene (PIB) as the sorbent film, which targets dimethyl methylphosphonate (DMMP), a simulant for nerve agent, sarin. An oscillator circuit was populated to trace the resonant frequency of the CMUT with a fast response time and low noise. We characterized the noise performance of the sensor system and identified the optimal gate time of the read-out frequency counter. Based on the noise measurement, the calculated limit of …

Background Low-intensity transcranial focused ultrasound stimulation is a promising candidate for noninvasive brain stimulation and accurate targeting of brain circuits because of its focusing capability and long penetration depth. However, achieving a sufficiently high spatial resolution to target small animal sub-regions is still challenging, especially in the axial direction. Objective To achieve high axial resolution, we designed a dual-crossed transducer system that achieved high spatial resolution in the axial direction without complex microfabrication, beamforming circuitry, and signal processing. Methods High axial resolution was achieved by crossing two ultrasound beams of commercially available piezoelectric curved transducers at the focal length of each transducer. After implementation of the fixture for the dual-crossed transducer system, three sets of in vivo animal experiments were conducted to demonstrate …

Distributed sensing of gas-phase chemicals using highly sensitive and inexpensive sensors is of great interest for many defense and consumer applications. In this paper we present ppb-level detection of dimethyl methylphosphonate (DMMP), a common simulant for sarin gas, with a ppt-level resolution using an improved capacitive micromachined ultrasonic transducer (CMUT) as a resonant chemical sensor. The improved CMUT operates at a higher resonant frequency of 47.7 MHz and offers an improved mass sensitivity of 48.8 zg/Hz/μm 2 by a factor of 2.7 compared to the previous CMUT sensors developed. A low-noise oscillator using the CMUT resonant sensor as the frequency-selective device was developed for real-time sensing, which exhibits an Allan deviation of 1.65 Hz (3σ) in the presence of a gas flow; this translates into a mass resolution of 80.5 zg/μm 2 . The CMUT resonant sensor is functionalized …

With the anticipation for a more connected world through the Internet of Things, there is still a strong demand for miniaturized chemical sensors. Here, we report on a miniaturized resonant chemical sensor based on a Capacitive Micromachined Ultrasonic Transducer (CMUT) with a low operating voltage suitable for portable gas sensor applications. Previously reported CMUT chemical sensors required a DC operating voltage (16∼50 V) higher than the supply voltages of common circuits ( e.g., 1.8–5 V). Thus, additional circuitry such as a charge pump circuit often was required as a part of the sensor interface circuits to supply the DC voltage to CMUT. This resulted in additional power consumption and a larger footprint. In this work, the vacuum gap of the CMUT which determines the operating voltage was reduced to 50 nm through the development of a double oxidation process with a high wafer-level yield. We …

Multi-functional neural probes are promising platforms to conduct efficient and effective in-depth studies of brain by recording neural signals as well as modulating the signals with various stimuli. Here we present a neural probe with an embedded microfluidic channel (chemtrode) with multi-drug delivery capability suitable for small animal experiments. We integrated a staggered herringbone mixer (SHM) in a 3-inlet microfluidic chip directly into our chemtrode. This chip, which also serves as a compact interface for the chemtrode, allows for efficient delivery of small volumes of multiple or concentration-modulated drugs via chaotic mixing. We demonstrated the successful infusion of combinatorial inputs of three chemicals with a low flow rate (170 nl min−1). By sequentially delivering red, green, and blue inks from each inlet and conducting visual inspections at the tip of the chemtrode, we measured a short residence …

With the rapid increase in the use of optogenetics to investigate the nervous system, there is a high demand for a neural interface that enables 2D mapping of electrophysiological neural signals with high precision during simultaneous light stimulation. Here, a gold nanonetwork (Au NN)‐based transparent neural electrocorticogram (ECoG) monitoring system is proposed as implantable neural electronics. The neural interface enables accurate 2D mapping of ECoG neural signals without any photoelectric artifact during light stimulation. By using the Au NN, not only the transmittance of the microelectrodes is increased by 81% but also a low electrochemical impedance of 33.9 kΩ at 1 kHz with improved mechanical stability is achieved. It is demonstrated that the transparent microelectrode array records multichannel in vivo neural activities with no photoelectric artifact and a high signal‐to‐noise ratio. Propagation of …

Distributed sensing of gas-phase chemicals is a promising application for mesoporous materials when combined with highly sensitive miniaturized gas sensors. We present a direct application of a mesoporous silica thin film on a highly sensitive miniaturized resonant chemical sensor with a mass sensitivity at the zeptogram scale for relative humidity and CO 2 detection. Using mesoporous silica thin-film, we report one of the lowest volume resolutions and a sensitive detection of 5.1 × 10 –4 % RH/Hz to water vapor in N 2 , which is 70 times higher than a device with a nontemplated silica layer. In addition, a mesoporous thin-film that is functionalized with an amino-group is directly applied on the resonant sensor, which exhibits a volume sensitivity of 1.6 × 10 –4 %/Hz and a volume resolution of 1.82 × 10 –4 % to CO 2 in N 2 .

Brain/Bio Medical Microsystems Lab

교수님 소개

이현주

연구실 소개

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Flexible ultrasound transducer array with statically adjustable curvature for anti-inflammatory treatment

Low-intensity focused ultrasound enables temporal modulation of human midbrain organoid differentiation

Ultrasound brain stimulation technologies for targeted therapeutics

연구실 보유 특허

대학별 입시끈

우리 연구실 모집

교수님 소개

이현주

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Flexible ultrasound transducer array with statically adjustable curvature for anti-inflammatory treatment

Low-intensity focused ultrasound enables temporal modulation of human midbrain organoid differentiation

Ultrasound brain stimulation technologies for targeted therapeutics

연구실 보유 특허

최근 3년간 연구실적 높은 연구실

대학별 입시끈

우리 연구실 모집