Introduction to VR surgical simualtion applications

Deukhee Lee(이득희) 대한기계학회 2018 대한기계학회 춘추학술대회 Vol.2018 No.12

Design and Implementation of Multimodal Simulation in Interactive Immersive Virtual Environment

Deukhee Lee,Sehyung Park,Sungdo Ha (사)한국CDE학회 2005 한국CAD/CAM학회 국제학술발표 논문집 Vol.2005 No.8

한 대의 카메라를 이용한 다점 3차원 위치 검출 알고리즘의 개발 : 보정된 카메라의 카메라 모델을 이용하여 크기와 색이 지정된 원형마커의 3차원 위치를 검출

이득희(Deukhee Lee),박세형(Sehyung Park) 한국HCI학회 2010 한국HCI학회 학술대회 Vol.2010 No.1

A Phantom Study on the Propagation of NIR Rays under the Skin for Designing a Novel Vein-Visualizing Device

Sangjun Lee,Sehyung Park,Deukhee Lee 제어로봇시스템학회 2013 제어로봇시스템학회 국제학술대회 논문집 Vol.2013 No.10

PCA와 kNN를 이용한 얼굴 인식률 향상을 위한 연구

강득희(DeukHee Kang),이말례(Marley Lee) 한국지능시스템학회 2008 한국지능시스템학회 학술발표 논문집 Vol.18 No.2

얼굴인식은 얼굴 영역의 분리, 특징 추출, 인식의 단계로 이루어져 있다. 얼굴인식의 주된 연구는 인식률을 향상을 목적으로 하고 있다, 이를 위해 본 논문은 인식률 향상을 위하여 다양한 테스트 영상에 대해 히스토그램 평활화(Histogram equalization)를 적용한 후 PCA(Principle Component Analysis)를 사용하여 고유벡터(Eigenvector)값을 단계별로 추출하여 다양한 크기의 kNN으로 인식률을 평가하는 얼굴 인식모델을 연구하였다. 본 논문에서 연구한 기법에 따라 실험을 통해 얻은 결과를 기반으로 높은 인식률을 가진 고유벡터값과 k값을 찾을 수 있었다.

Intervertebral disc instance segmentation using a multistage optimization mask-RCNN (MOM-RCNN)

Vania Malinda,Lee Deukhee 한국CDE학회 2021 Journal of computational design and engineering Vol.8 No.4

Lower back pain is one of the major global challenges in health problems. Medical imaging is rapidly taking a predominant position for the diagnosis and treatment of lower back abnormalities. Magnetic resonance imaging (MRI) is a primary tool for detecting anatomical and functional abnormalities in the intervertebral disc (IVD) and provides valuable data for both diagnosis and research. Deep learning methods perform well in computer visioning when labeled general image training data are abundant. In the practice of medical images, the labeled data or the segmentation data are produced manually. However, manual medical image segmentation leads to two main issues: much time is needed for delineation, and reproducibility is called into question. To handle this problem, we developed an automated approach for IVD instance segmentation that can utilize T1 and T2 images during this study to handle data limitation problems and computational time problems and improve the generalization of the algorithm. This method builds upon mask-RCNN; we proposed a multistage optimization mask-RCNN (MOM-RCNN) for deep learning segmentation networks. We used a multi-optimization training system by utilizing stochastic gradient descent and adaptive moment estimation (Adam) with T1 and T2 in MOM-RCNN. The proposed method showed a significant improvement in processing time and segmentation results compared to previous commonly used segmentation methods. We evaluated the results using several different key performance measures. We obtain the Dice coefficient (99%). Our method can define the IVD’s segmentation as much as 88% (sensitivity) and recognize the non-IVD as much as 98% (specificity). The results also obtained increasing precision (92%) with a low global consistency error (0.03), approaching 0 (the best possible score). On the spatial distance measures, the results show a promising reduction from 0.407 ± 0.067 mm in root mean square error to 0.095 ± 0.026 mm, Hausdorff distance from 12.313 ± 3.015 to 5.155 ± 1.561 mm, and average symmetric surface distance from 1.944 ± 0.850 to 0.49 ± 0.23 mm compared to other state-of-the-art methods. We used MRI images from 263 patients to demonstrate the efficiency of our proposed method.

Registration accuracy enhancement of a surgical navigation system for anterior cruciate ligament reconstruction: A phantom and cadaveric study

Kim, Youngjun,Lee, Byung Hoon,Mekuria, Kinde,Cho, Hyunchul,Park, Sehyung,Wang, Joon Ho,Lee, Deukhee Elsevier 2017 The knee Vol.24 No.2

<P><B>Abstract</B></P> <P><B>Background</B></P> <P>Recently, surgical navigation systems have been widely used to improve the results of various orthopaedic surgeries. However, surgical navigation has not been successful in anterior cruciate ligament reconstruction, owing to its inaccuracy and inconvenience. This study investigated the registration of preoperative and intraoperative data, which are the key components in improving accuracy of the navigation system.</P> <P><B>Methods</B></P> <P>An accurate registration method was proposed using new optical tracking markers and landmark retake. A surgical planning and navigation system for anterior cruciate ligament reconstruction was developed and implemented. The accuracy of the proposed system has been evaluated using phantoms and eight cadaveric knees. The present study investigated only the registration accuracy excluding the errors of optical tracking hardware and surgeon.</P> <P><B>Results</B></P> <P>The target registration errors of femoral tunnelling for anterior cruciate ligament reconstruction in phantoms were found to be 0.24±0.03mm and 0.19±0.10° for the tunnel entry position and tunnel direction, respectively. The target registration errors measured using cadavers were 0.9mm and 1.94°, respectively.</P> <P><B>Conclusions</B></P> <P>The preclinical experimental results showed that the proposed methods enhanced the registration accuracy of the developed system. As the system becomes more accurate, surgeons could more precisely position and orient the femoral and tibial tunnels to their original anatomical locations.</P>

Image-guided handheld HIFU treatment system based on real-time tracking of ultrasound imaging probe and HIFU probe

Kyusic Park,Deukhee Lee 제어로봇시스템학회 2013 제어로봇시스템학회 국제학술대회 논문집 Vol.2013 No.10

뇌 내출혈종 수술을 위한 가중 ICP 알고리즘 기반 마커리스 환자-영상 정합 방법

서준호(Joonho Seo),이득희(Deukhee Lee),김영준(Youngjun Kim),박세형(Sehyung Park) (사)한국CDE학회 2012 한국 CAD/CAM 학회 학술발표회 논문집 Vol.2012 No.2

In the conventional Intracerebral Hematoma (ICH) removal surgery, it has been required to install the stereotactic frame on the patient’s cranial surface to access to the intracerebral hematoma. Since it is invasive and also takes long time, we attempt to develop a robotic ICH removal procedure with a markerless registration applying optical 3-D scanner. Preoperative patient’s data should be investigated from the CT images; it includes patient’s 3-D facial surface, hematoma, and surgical plans. Intraoperatively scanned facial surface will be registered to navigate the hematoma in the current pose of patient. Here, the conventional ICP (Iterative Closest Point) algorithm can be used for the registration. However, we need to assume that the patient’s face should not be changed from the preoperative model since the ICP considers only rigid-body transformation. In this paper, therefore, we investigated some facial features, which can be used as anatomical landmarks to identify the position of hematoma in the cranial structure. Moreover, the facial features in the preoperative 3-D model are weighted for more accurate registration with intraoperatively scanned points, ignoring locally deformed facial shapes. Accordingly, more accurate and robust target identification can be achieved even with patient’s local facial shape changes from the preoperative data.

불안정성 원위 요골 골절의 치료에 있어 한국형 이중 연골하 지지고정 전방 금속판의 임상적 및 방사선학적 결과

이철형(Chul-Hyung Lee),정덕희(Deukhee Jung),안충한(Chung-Han An),정의탁(Uitak Jeong) 대한정형외과학회 2020 대한정형외과학회지 Vol.55 No.6

목적: 본 연구는 요골 원위부 골절에 대해 국내에서 이중 연골하 지지고정 개념으로 개발된 전방 잠김 금속판의 수술 후 정복 유지효과 및 이에 영향을 줄 수 있는 요인들에 대해 평가해 보고자 하였다. 대상 및 방법: 2017년 7월부터 2018년 12월까지 요골 원위부 골절 환자 중 관혈적 정복 및 금속판 고정술을 시행한 54예를 대상으로 하였다. 수술 직후와 최종 추시 시 촬영한 단순방사선 사진을 이용해 요골 길이, 요골경사, 수장측 경사, 척골 변위, 원위 배측 피질 거리(distal dorsal cortical distance)를 측정하여 골절 정복 유지 효과에 대해 평가하였다. 환자 나이, 골절 분류, 금속판 위치를 기준으로 전체 환자군을 세부 그룹으로 나누어 비교함으로써 각 요인이 골절 정복 유지에 미치는 영향을 분석하였다. 결과: 수술 직후 원위 배측 피질 거리는 평균 5.91 mm (표준 편차, ±1.95 mm)로 측정되었으며 요골 길이(p=0.038), 척골 변위(p=0.001)는 수술 직후와 최종 추시 시 촬영한 단순 방사선 사진에서 유의한 차이가 확인되었다. 전체 환자군을 각 요인에 따라 세부 그룹으로 나누어 평가했을 시 척골 변위는 나이가 65세 이상이거나 AO/OTA 분류 C3형에 해당하는 경우 혹은 금속판 위치가 Soong classification grade 0에 해당하는 경우 수술 직후에 비해 최종 추시 시 유의하게 증가하였다(p=0.007, p=0.012, =p0.046). 결론: 국내에서 이중 연골하 지지고정 개념으로 생산된 전방 잠김 금속판을 이용하여 요골 원위부 골절 정복술을 시행하는 경우 충분한 정복을 확보할 수 있으며 골절 정복 유지에 대해 전반적으로 긍정적인 결과를 보이나, 환자가 고령인 경우, 관절 내 골절의 분쇄 정도가 심한 경우, 금속판이 분수령선의 근위부에 위치한 경우에 척골 변위의 증가를 보여 요골 월상골와의 정복 유지를 위한 고려가 필요하다. Purpose: The aim of this study was to assess the effectiveness of domestically developed volar locking plate which has the concept of double-tiered subchondral support (DSS) in maintaining the reduction after distal radial fracture surgery. Materials and Methods: From July 2017 to December 2018, 54 patients were assessed. Plain radiographs were obtained immediately after surgery and at the last follow-up, and the radiographic parameters were measured in those images: radial length, radial inclination, volar tilt, ulnar variance, and distal dorsal cortical distance. The patients were subdivided into their age, type of fracture, and the position of the plate to evaluate the influence of each factors on the reduction maintenance. Results: Distal dorsal cortical distance in radiographs after the surgery was 5.91 mm (standard deviation, ±1.95 mm) on average. Significant differences in the radial length (p=0.038) and ulnar variance (p=0.001) were observed between immediately after surgery and at the last follow-up. When the parameters were evaluated by dividing the patients into subgroups according to the three specific factors, the ulnar variance showed a significant increase at the last follow-up when the patients were included 65-years-old or older. AO/OTA type C3 fracture, and Soong classification grade 0 plate position (p=0.007, p=0.012, p=0.046, respectively). Conclusion: Using the domestically developed DSS-type volar locking plate, significant reduction after distal radial fracture surgery could be maintained successfully. On the other hand, further study will be needed to determine about the reduction loss of the lunate facet identified in special cases that deal with fractures in elderly patients, unstable AO/OTA type C3 distal radial fractures, and Soong classification grade 0 plate position.