Research

aicd
이동: 둘러보기, 검색

연구 프로젝트[편집]

진행중인 프로젝트[편집]

  • Deep Health Eye: 시각지능을 통한 인체의 3대 생체정보 (맥박, 산소포화도, 혈압) 획득 및 이를 활용한 복합감정 및 건강상태 예측 시스템 연구 (2017-2022), 한국연구재단 이공학 개인기초연구지원사업 - 책임
  • 뇌유래성장인자 발현을 증가시키는 LED 광파장에 의한 기억증진 효능 확인 (2017-2020), 한국연구재단 이공학 개인기초연구지원사업 - 책임 ( AICD 허진철 박사)
  • 광간섭성 단층촬영을 사용한 혈관 내 단면 영상에서 다양한 병변에 대한 인공지능 기반 시각적 자동분석 및 컴퓨터 보조 진단 시스템 개발 (2017-2021), 한국연구재단 이공학 개인기초연구지원사업 - 책임 (동산의료원 신경외과 김창현 교수)
  • 차세대 뇌혈관 수술을 위한 광간섭성 단층촬영용 브레인 카테터 개발 (2017-2018), 대구 의료분야 연구자 창업 지원사업, DGMIF 첨단의료기기개발지원센터 - 책임 (동산의료원 신경외과 김창현 교수)
  • 인공지능 기반 시각정보를 통한 생체정보 (맥박, 산소포화도, 혈압) 획득 및 이를 활용한 복합감정 및 건강상태 예측 시스템 연구 (2017-2020), 계명스칼라연구비 - 책임
  • 실생활 기반 사용성평가 지원센터 구축사업 (2015-2020), 산업기술거점기관지원사업, 산업통상자원부 - 총괄 책임
  • 심혈관질환 첨단의료기술 가상훈련시스템 기술개발 (2017-2021), 보건의료기술연구개발사업, 보건복지부 - 위탁
  • 유방암 초음파 영상판독 보조진단 소프트웨어 및 유사환자 검색시스템 개발 (2017-2019), 지역특화산업육성사업, 산업통상자원부 - 참여기관 책임
  • 자기 주도형 휴대용 생활환경 안전진단 키트 및 앱기반 서비스 시스템 (2016-2020), 산업기술혁신사업, 산업통상자원부 - 참여기관 책임
  • 90% 이상의 정확도의 생체신호 측정과 Biofeedback을 통한 IoT기반 스트레스 조절 기능의 사용자 감성충족 디자인 적용 여성용 스마트 의류일체형 디바이스 개발 (2017-2019), 산업기술혁신사업, 산업통상자원부 - 참여기관 참여
  • 의료기기 표준 플랫폼 기술개발 및 보급 활성화 지원 (2016-2021), 산업기술기반구축사업, 산업통상자원부 - 주관기관 참여

수행완료 프로젝트[편집]

  • [종료] 임베디드SW 전문인력양성사업 (2016-2020), 산업전문인력역량강화사업, 산업통상자원부 - 참여기관 참여
  • [종료] 인체활동 통합관리지원을 위한 다중 웨어러블 SW융합모듈 및 유연 SW 응용플랫폼 기술개발 (2016-2019), 정보통신·방송 연구개발사업, 미래창조과학부 (MSIP) - 참여기관 참여
  • [종료] 거치형 하지 재활훈련과 이동형 능동 보행을 함께 지원하는 하이브리드 하지재활시스템 (2016-2017), 경제협력권산업육성 기술개발사업, 산업통상자원부 - 참여기관 책임
  • [종료] 광학 영상을 이용한 초저가 유방암 조기 진단 기기 개발 (2015-2017), 연구개발특구육성사업, 연구개발특구진흥재단 - 참여기관 책임
  • [종료] 고령친화제품 사용성평가 전문기관 시범 운영 사업 (2017-2018), 한국보건산업진흥원 - 연구용역 참여
  • [종료] VC 유형별 사용성평가 분석 연구 용역 (2016-2017), (주)삼성전자 - 용역기관 참여
  • [종료] 유방암 초기증상의 자가진단을 정량화하고 병변을 진단/예측하기 위한 촉각 영상화 방법 연구 (2014-2017), 일반연구자지원사업, 한국연구재단 - 연구 책임
  • [종료] 사용자의 체온을 유지, 상승 및 하강시킬 수 있는 웨어러블 체온 유도 디바이스 개발 (2016-2017), 산학연협력기술개발사업, 중소기업청 - 연구 책임
  • [종료] 비타민D 측정기가 부착된 가정용 UVB 기기 개발 (2016-2017), 산학연협력기술개발사업, 중소기업청 - 연구 책임
  • [종료] 만성질환 관리를 위한 인체삽입형 생리기능 자동감시 시스템 기술 (2012-2017), 정보통신·방송 연구개발사업, 미래창조과학부 (MSIP) - 참여기관 참여
  • [종료] 지속적인 전력 공급을 위한 초박형 피부접착 생체센서용 무선전력전송 모듈 개발(2016-2017), 산학공동기술개발과제, 한국연구재단 - 연구 책임
  • [종료] 혈류량 감지 시스템 기술개발 (2014-2016), 기술혁신개발사업, 중소기업청 - 연구 책임
  • [종료] 4060세대 심질환자 관리를 위한 90% 이상의 정확도를 가지는 부정맥 자동 검출 센서 장착 아웃도어용 스마트 셔츠 상용화 기술개발 (2015-2016), 글로벌전문기술개발사업, 산업통상자원부 - 참여기관 참여
  • [종료] 수요연계형 Daily Healthcare 실증단지 조성사업 (2015-2018), 미래창조과학부 (MSIP) - 참여기관 참여
  • [종료] 지속적인 생체모니터링을 위한 접착형 인공피부 Epidermal Electronics 센서 개발 (2015), (재)동일문화장학재단 - 연구 책임
  • [종료] 빅데이터 기반 초음파 영상 자동 진단 소프트웨어 개발 (2015-2016), 산학공동기술개발과제, 한국연구재단 - 연구 책임
  • [종료] 의료기기, 의료로봇 기술 개발 및 의료IT융합 활성화 사업 (빅데이터 기반 인공지능 컴퓨터 통합 진단 솔루션) (2015), 한국전자통신연구원 (ETRI) - 연구 책임
  • [종료] 심혈관계 질환 의료기술 훈련 역량강화 R&D 기획 (2015-2016), 대구경북첨단의료산업진흥재단, 첨단의료기기개발지원센터 - 연구 책임
  • [종료] 일반인(직장인)의 균형 잡힌 웰니스 증진을 위한 응용 서비스 플랫폼 구축 (2013-2016), 미래산업선도기술개발사업, 산업통상자원부 - 연구 책임
  • [종료] 삼차원 멀티 터치스크린 시스템용 의료영상 컨텐츠 솔루션 개발 (2015), (주)알앤디플러스 - 연구 책임
  • [종료] 촉감 영상을 이용한 초저가 유방암 조기 진단 기기 개발 (2014-2015), 산학공동기술개발과제, 한국연구재단 - 연구 책임
  • [종료] 심장 진환의 진단 편의성을 위한 대용량 심전도 신호 자동 분석 시스템 개발 (2013-2014), 산학공동기술개발과제, 한국연구재단 - 연구 책임
  • [종료] 유방암 조기 발견을 위한 초음파 영상 자동 진단 시스템 개발 (2013-2014), (재)산학협동재단 - 연구 책임
  • [종료] 생체 신호 분석을 통한 부정맥 진단 알고리즘 및 모니터링 리포트 시스템 개발 (2013-2014), 산학연협력 기술개발 지원사업, 중소기업청 - 연구 책임
  • [종료] 부정맥 질환 진단/치료기기 개발 및 상용화 지원사업 (2013-2015), 광역경제권연계협력사업, 산업통상자원부 - 참여기관 참여

Optogenetics[편집]

Optogenetics is a biological technique which involves the use of light to control cells in living tissue, typically neurons, that have been genetically modified to express light-sensitive ion channels. It is a neuromodulation method employed in neuroscience that uses a combination of techniques from optics and genetics to control and monitor the activities of individual neurons in living tissue—even within freely-moving animals—and to precisely measure the effects of those manipulations in real-time. The key reagents used in optogenetics are light-sensitive proteins. Neuronal control is achieved using optogenetic actuators like channelrhodopsin, halorhodopsin, and archaerhodopsin, while optical recording of neuronal activities can be made with the help of optogenetic sensors for calcium (GCaMP), vesicular release (synaptopHluorin), neurotransmitter (GluSnFRs), or membrane voltage (Arclightning, ASAP1). Control or recording is confined to genetically defined neurons and performed in a spatiotemporally precise manner by light.

Opto5.png

High-Resolution Tactile Imaging Sensor Using Total Internal Reflection and Nonrigid Pattern Matching Algorithm[편집]

This is a development of comparative studies of elasticity using optical based high resolution elasticity imaging. Total internal reflection based elasticity imaging device and forward/inversion based stress/strain measurement algorithm are developed to extract fully elasticity map of the tissue. This is the first validation study of optical based elasticity estimation for tumor detection with direct comparison based on in vivo data. The tissue inclusion parameter estimation method is proposed to measure the stiffness as well as geometric parameters. The estimation is performed based on the tactile data obtained at the surface of the tissue using an optical tactile sensation imaging system (TSIS). A forward algorithm is designed to comprehensively predict the tactile data based on the mechanical properties of tissue inclusion using finite element modeling (FEM). This forward information is used to develop an inversion algorithm that will be used to extract the size, depth, and Young's modulus of a tissue inclusion from the tactile data. We utilize the artificial neural network (ANN) for inversion algorithm. The proposed estimation method was validated by the realistic tissue phantom with stiff inclusions.

Tsi2.png

Human Implantable Arrhythmia Monitoring Sensor with Wireless Power and Data Transmission Technique[편집]

Prolonged monitoring is more likely to diagnose atrial fibrillation accurately than intermittent or short-term monitoring. In this study, an implantable electrocardiograph (ECG) sensor to monitor atrial fibrillation patients in real time was developed. The implantable sensor is composed of a micro controller unit, an analog-to-digital converter, a signal transmitter, an antenna, and two electrodes. The sensor detects ECG signals from the two electrodes and transmits these to an external receiver carried by the patient. Because the sensor continuously transmits signals, its battery consumption rate is extremely high; therefore, the sensor includes a wireless power transmission module that allows it to charge wirelessly from an external power source. The integrated sensor has the approximate dimensions 0.12 in x 1.18 in x 0.19 in, which is small enough to be inserted into a patient without the need for major surgery. The signal and power transmission data sampling rate and frequency of the unit are 300 samples/s and 430 Hz, respectively. To validate the developed sensor, experiments were conducted on small animals.

Bio1.png

Deep Learning Based Computer Aided Diagnostics[편집]

Early detection of breast cancer is critical in determining the best possible treatment approach. Due to its superiority compared with mammography in its ability to detect focal abnormalities in dense breast tissue, ultrasound imaging has become an important modality in breast tumor detection and classification. This paper discusses the novel Fourier-based shape feature extraction techniques that provide enhanced classification accuracy for breast tumor in the computer-aided B-mode ultrasound diagnosis system. To verify the effectiveness of the proposed features, experiments were performed using 4,107 ultrasound images containing 2,508 malignancy cases. Experimental results show that the breast tumor classification accuracy of the proposed technique was 15.8%, 5.43%, 17.32%, and 13.86% higher than the previous shape features such as number of protuberances, number of depressions, lobulation index, and dissimilarity, respectively.

Big4.jpg

Terahertz Radiation Imaging[편집]

Terahertz imaging is an emerging and significant nondestructive evaluation (NDE) technique used for dielectric (nonconducting, i.e., an insulator) materials analysis and quality control in the pharmaceutical, biomedical, security, materials characterization, and aerospace industries. It has proved to be effective in the inspection of layers in paints and coatings, detecting structural defects in ceramic and composite materials and imaging the physical structure of paintings and manuscripts. The use of THz waves for non-destructive evaluation enables inspection of multi-layered structures and can identify abnormalities from foreign material inclusions, disbond and delamination, mechanical impact damage, heat damage, and water or hydraulic fluid ingression. This new method can play a significant role in a number of industries for materials characterization applications where precision thickness mapping (to assure product dimensional tolerances within product and from product-to-product) and density mapping (to assure product quality within product and from product-to-product) are required.

Tera11.jpg

Non-Contact Bio Signal Measurement[편집]

Nowadays, since the rapid development of the wearable device, bio signal measurement becomes easier and simpler. However, the wearable devices are contact to person method. In this paper, the non-contact method to measure photoplethsmograph (PPG) and pulse oximetry using face image sequences. By using the image sequences, user can measure bio-signals without sensors. First, subtle motion which heartbeat occurs in video was amplified to get PPG using Eulerian motion magnification. From the video, the image was magnified frame by frame to see the slight difference of skin variation occurred by the blood flows. Next, pulse oximetry was measured from comparing the oxygen saturated hemoglobin and oxygen unsaturated hemoglobin in the blood. Two light sources in 765nm and 880nm wavelength alternately illuminated user skin. Then mobile camera captured lights reflected from the skin.

Biosignal.jpg


Stabilization of Aging Neurons Through Activation of Hippocampal BDNF by Optical Method[편집]

BDNF (Brain-derived neurotrophic factor) acts on specific neurons in the central nervous system and the peripheral nervous system to promote survival of neurons, growth and differentiation of new neurons and synapses, involved in learning, long-term memory, and thinking. Some of the adult brain may have the capacity to grow new neurons from neural stem cells in the process known as neurogenesis. BDNF plays an important role in normal neural development. The team is working on the development of devices capable of expressing/secreting BDNF in the hippocampus, directly or indirectly, using optical wavelengths.

Bbbb.png

Development of Early Diagnosis and Treatment System for Atopic Dermatitis using Terahertz Wave[편집]

In this study, we investigated the diagnosis and treatment methods of high-quality terahertz for atopic dermatosis, and want to study diagnostic and treatment systems. Terahertz spectroscopy and disease model image analysis algorithm optimized for atopic dermatosis diagnosis, computer aided diagnosis algorithm based on artificial intelligence and integrated diagnosis of terahertz image and user data to suggest optimal diagnosis and treatment strategy, verification of in vivo atopic dermatosis diagnosis and treatment efficacy using terahertz and performance evaluation of artificial intelligence computer assisted diagnosis algorithm. This system propose a new approach to the development of scientific and standardized diagnostic methods for diagnostic methods give a biased visual information.

Aaaa.png


Computer Aided Diagnosis for Intravascular Optical Coherence Tomography images[편집]

Intravascular Optical Coherence Tomography (IVOCT) is a medical imaging modality, which provides high images resolution (10-15 μm) cross sectional image of coronary artery. Therefore, IVOCT allows evaluation of thrombus and plaque. Clinically, irregular protrusions and thrombosis after stent deployment can be significant adverse outcomes such as thrombotic re-occlusion or restenosis. In IVOCT images, if thrombosis is present in the blood vessel, the vessel lumen has irregular shape feature and specific texture feature. In this study, we propose an algorithm for automatically detect lumen morphology and analyze lumen shape feature and texture feature. Then, extracted features were classified into normal/abnormal lumen using machine learning technique. 3-dimensional reconstruction of the lumen was performed to intuitively read the morphological shape of the lumen. In the 3-dimensional lumen model, normal/abnormal lumen are depicted with different colors to visualize irregularity of lumen. In the results, we noticed that the normal/abnormal lumen frame and location of the thrombosis in 3-dimensional lumen can be classified and analyzed in a few seconds and can lead to understanding of overall vascular status and help to determine cardiovascular diagnosis.

IVOCT2.png