• 대한의용생체공학회:의공학회지
• /
• 제43권4호
• /
• pp.259-270
• /
• 2022

Artificial intelligence (AI) has been studied in various fields of medical imaging. Currently, top-notch deep learning (DL) techniques have led to high diagnostic accuracy and fast computation. However, they are rarely used in real clinical practices because of a lack of reliability concerning their results. Most DL models can achieve high performance by extracting features from large volumes of data. However, increasing model complexity and nonlinearity turn such models into black boxes that are seldom accessible, interpretable, and transparent. As a result, scientific interest in the field of explainable artificial intelligence (XAI) is gradually emerging. This study aims to review diverse XAI approaches currently exploited in medical imaging. We identify the concepts of the methods, introduce studies applying them to imaging modalities such as computational tomography (CT), magnetic resonance imaging (MRI), and endoscopy, and lastly discuss limitations and challenges faced by XAI for future studies.

• Journal of the Optical Society of Korea
• /
• 제15권3호
• /
• pp.300-304
• /
• 2011

Optical microscopes are widely used for medical imaging these days, but biopsy is a lengthy process that causes many problems during the ex-vivo imaging procedure. The endo-microscope has been studied to increase accessibility to the human body and to get in-vivo images to use for medical diagnosis. This research proposes a multi-modal confocal endo-microscope for bio-medical imaging. We introduce the design process for a small endoscopic probe and a coupling mechanism for the probe to make the multi-modal confocal endo-microscope. The endoscopic probe was designed to decrease chromatic and spherical aberrations, which deteriorate the images obtained with the conventional GRIN lens. Fluorescence and reflectance images of various samples were obtained with the proposed endo-microscope. We evaluated the performance of the proposed endo-microscope by analyzing the acquired images, and demonstrate the possibilities of in-vivo medical imaging for early diagnosis.

• 한국의학물리학회:학술대회논문집
• /
• 한국의학물리학회 2006년도 제33회 추계학술대회 발표논문집
• /
• pp.156-156
• /
• 2006

• 대한의용생체공학회:의공학회지
• /
• 제26권5호
• /
• pp.283-294
• /
• 2005

The Center for Imaging Human Structures (CIH) was established in December 2002 to develop new diagnostic imaging techniques and to make them available to the greater community of biomedical and clinical researchers at Sungkyunkwan University. CIH has been involved in 5 specific activities to provide solutions for early diagnosis and improved treatment of human diseases. The five area goals include: 1) development of a digital mammography system with computer aided diagnosis (CAD); 2) development of digital radiological imaging techniques; 3) development of unified medical solutions using 3D image fusion; 4) development of multi-purpose digital endoscopy; and, 5) evaluation of new imaging systems for clinical application

• 대한의용생체공학회:의공학회지
• /
• 제44권4호
• /
• pp.255-263
• /
• 2023

Clinical ultrasound (US) is a widely used imaging modality with various clinical applications. However, capturing a large field of view often requires specialized transducers which have limitations for specific clinical scenarios. Panoramic imaging offers an alternative approach by sequentially aligning image sections acquired from freehand sweeps using a standard transducer. To reconstruct a 3D volume from these 2D sections, an external device can be employed to track the transducer's motion accurately. However, the presence of optical or electrical interferences in a clinical setting often leads to incorrect measurements from such sensors. In this paper, we propose a deep learning (DL) framework that enables the prediction of scan trajectories using only US data, eliminating the need for an external tracking device. Our approach incorporates diverse data types, including correlation volume, optical flow, B-mode images, and rawer data (IQ data). We develop a DL network capable of effectively handling these data types and introduce an attention technique to emphasize crucial local areas for precise trajectory prediction. Through extensive experimentation, we demonstrate the superiority of our proposed method over other DL-based approaches in terms of long trajectory prediction performance. Our findings highlight the potential of employing DL techniques for trajectory estimation in clinical ultrasound, offering a promising alternative for panoramic imaging.

• 인터넷정보학회논문지
• /
• 제16권2호
• /
• pp.109-115
• /
• 2015

신규 의료기기 개발 기술의 발전으로 다양한 형태로 손쉽게 생체 정보 및 의료 정보를 얻을 수 있는 기술이 증가하고 있다. 이러한 정보 수집 기술과 기기들의 증가로 생체 정보는 일상생활의 라이프로그와 함께 의료서비스의 주요 정보로 활용되고 있다. 그러나 다양한 생체신호의 활용성이 증가하고 있지만 보안적인 측면을 고려하지 않는 문제점을 갖고 있다. 또한, 의료현장에서 환자의 생체신호와 의료영상정보는 개별적인 디바이스에 의해 생성되며, 통합 관리되지 못하는 실정이다. 이러한 문제점을 해결하기 위해서, 본 논문에서는 생체신호와 의사의 소견정보를 포함하여 QR 코드화하고 이와 연계된 의료영상정보와 통합하고자 한다. 이를 위해, 의료영상정보 표준인 DICOM(Digital Imaging and Communication in Medicine)과 기존 생체신호 계측기들로부터 수집된 생체신호를 QR 코드화하여 의료영상정보에 통합한 이미지 파일 스킴을 제시한다. 그리고 시스템 구현 환경은 의료영상기기와 생체신호 수집을 위한 생체신호 계측기 그리고 스마트 디바이스와 PC로 구성하였다. 의료기기나 생체 신호 계측장치로부터 데이터를 전송 받기 위한 의료영상이미지 정보와 생체신호의 ROI 추출을 위하여 .NET Framework를 사용하여 QR 서버 모듈을 윈도우 서버 2008 운영체제에서 운영되도록 구현하였다. QR 서버 모듈의 주요기능은 의료영상기기로부터 생성된 DICOM파일을 파싱하고, 식별 ROI 정보를 추출하여 데이터베이스에 저장하여 관리한다. 또한, EMR, OCS와 같은 환자의 의료정보는 기본 정보 및 긴급상황 시 필요한 ROI 정보를 추출하여 QR코드화 하여 관리한다. 또한 생체 계측 기기로 환자 식별에 사용될 PID (patient identification) 와 함께 생체 정보를 전송 받을 경우 생체 정보의 크기에 따라 이를 해당 환자의 ROI와 함께 QR코드화 하여 관리하며, 생체 정보 파일 또한 저장하여 관리한다. 전송받은 생체정보가 QR코드로 변환할 최대 사이즈 이상일 경우 서버를 통해 생체정보에 접근할 수 있는 URL 정보를 QR코드화 한다. 또한 QR 코드 형태로 제공되는 정보는 .NET 프레임워크가 설치된 PC와 Android기반의 스마트 단말기상에 뷰어 프로그램을 통해 확인함으로 인증된 클라이언트만이 관련 정보를 확인할 수 있도록 하였다. 끝으로 응용 서비스의 수행결과를 통해 기존 의료영상정보와 생체신호 그리고 환자의 건강정보가 통합되어 의료현장에서 적용하는데 적합한 의료정보 서비스를 제공함을 보였다.

• 대한의용생체공학회:의공학회지
• /
• 제40권6호
• /
• pp.230-234
• /
• 2019

Medical laser equipment using optical energy is used to surgery and treat diseases by destroying and removing tissue. Domestic laser equipment has been used steadily in the skin and cosmetics sectors and has been changed to radiate high-power energy in a wide range to shorten patient treatment time. However, side effects such as burns and damage of normal tissues occurred. To solve this problem, techniques for detecting lesions using an imaging device and selectively radiating the laser have been developed. In this study, we proposed an evaluation method to evaluate the safety and performance of target detection accuracy, laser irradiation accuracy and motion protection device technology derived from product analysis and investigation. Finally, the validity of the evaluation method was evaluated by evaluating the imaging device based laser equipment as the proposed evaluation method.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2010년도 춘계학술발표논문집 2부
• /
• pp.1227-1229
• /
• 2010

This paper explains what DTI (Diffusion Tensor Imaging) is and what it does. We will talk about the DTI functions and what type of image it can show, and what areas are using DTI. The tractography and other applications that DTI is being used. In this paper, we explain that DTI is not only useful in medicine but also in botany. We propose to use DTI to study structure and functions of plants.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2010년도 제39회 하계학술대회 초록집
• /
• pp.73-73
• /
• 2010

Various organic- and bio-conjugated nanoparticles have been studied extensively for biological applications in medical diagnoses and drug delivery systems. Gold nanoparticles (AuNP) and poly(ethylene glycol) (PEG) are known biocompatible materials to be used in vivo and are becoming increasingly important in biomedical applications. In this work, we investigated the stability of PEG-conjugated AuNPs, dialysis and centrifuge effects after synthesis or pegylation of AuNPs as a function of elapsed time by using ToF-SIMS imaging technique along with dynamic light scattering (DLS), UV-visible absorption spectroscopic and statistical analyses. Roughly 15-nm-sized AuNPs were synthesized in a citrate-conjugated form, and then converted into the thiol-terminated PEG (O-[2-(3-Mercaptopropionylamino)ethyl]-O'-methylpolyethyleneglycol, M.W.=5 kDa) form. Based on our data, we will show that ToF-SIMS imaging analysis along with DLS, UV-visible absorption and statistical analyses would be a useful method to evaluate stability of PEG-conjugated AuNPs in various environmental conditions.

• 생명과학회지
• /
• 제20권11호
• /
• pp.1738-1741
• /
• 2010

기존의 동물모델에서 암의 성장은 caliper를 이용하여 고형암 부피를 측정으로써 조사하였으나, 암 조직 속의괴사와 부종으로 인하여 부피측정에 신뢰성이 결여 되어 있다. 이러한 문제점을 해결하기 위해 발광 암세포를 이용하여 광학생체영상적으로 분석하는 방법이 개발 되었다. 본 연구에서는 전립선 발광 암세포를 제조하여 고형암 동물모델에서 B16 발광 암세포와 암 성장을 비교 측정하여 신규발광 암세포를 평가하였다. In vitro에서 세포 수와 발광강도는 높은 상관관계를 보였고($R^2$=0.99), 고형암 동물모델에서 암 성장 측정은 괴사에 의한 오차를 줄였다. 이러한 발광신호를 기반으로 한 측정방법은 caliper의 부피 측정에 비하여 높은 항암효과를 보임으로써 기존의 발광 암세포보다 신규 발광전립선 암세포의 유용성을 증명하였다.