Sungon Lee

Dr. Sungon Lee

Associate Professor, School of Electrical Engineering, Hanyang University, Ansan 15588, Korea
orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 February 2019 in Applied Sciences
Verified
Journal Article
Open access
PDF file
HHHyunho Hwang
Hyunho Hwang
HHafiz Zia Ur Rehman
Hafiz Zia Ur Rehman
Sungon LeeSungon Lee
Sungon Lee
Skull stripping in brain magnetic resonance imaging (MRI) is an essential step to analyze images of the brain. Although manual segmentation has the highest accuracy, it is a time-consuming task. Therefore, various automatic segmentation algorithms of the brain in MRI have been devised and proposed previously. However, there is still no method that solves the entire brain extraction problem satisfactorily for diverse datasets in a generic and robust way. To address these shortcomings of existing methods, we propose the use of a 3D-UNet for skull stripping in brain MRI. The 3D-UNet was recently proposed and has been widely used for volumetric segmentation in medical images due to its outstanding performance. It is an extended version of the previously proposed 2D-UNet, which is based on a deep learning network, specifically, the convolutional neural network. We evaluated 3D-UNet skull-stripping using a publicly available brain MRI dataset and compared the results with three existing methods (BSE, ROBEX, and Kleesiek’s method; BSE and ROBEX are two conventional methods, and Kleesiek’s method is based on deep learning). The 3D-UNet outperforms two typical methods and shows comparable results with the specific deep learning-based algorithm, exhibiting a mean Dice coefficient of 0.9903, a sensitivity of 0.9853, and a specificity of 0.9953.

4 March 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
HHafiz Zia Ur Rehman
Hafiz Zia Ur Rehman
HHHyunho Hwang
Hyunho Hwang
Sungon LeeSungon Lee
Sungon Lee
Skull stripping in brain magnetic resonance volume has recently been attracting attention due to an increased demand to develop an efficient, accurate, and general algorithm for diverse datasets of the brain. Accurate skull stripping is a critical step for neuroimaging diagnostic systems because neither the inclusion of non-brain tissues nor removal of brain parts can be corrected in subsequent steps, which results in unfixed error through subsequent analysis. The objective of this review article is to give a comprehensive overview of skull stripping approaches, including recent deep learning-based approaches. In this paper, the current methods of skull stripping have been divided into two distinct groups—conventional or classical approaches, and convolutional neural networks or deep learning approaches. The potentials of several methods are emphasized because they can be applied to standard clinical imaging protocols. Finally, current trends and future developments are addressed giving special attention to recent deep learning algorithms.

20 November 2018 in IEEE Access
Verified
Journal Article
Open access
HHafiz Zia Ur Rehman
Hafiz Zia Ur Rehman
Sungon LeeSungon Lee
Sungon Lee
We present an automatic technique for image alignment using principal component analysis (PCA) that broadly consists of two steps. The first step is the segmentation of the region of interest by thresholding. In the second step, PCA is applied on nonzero pixels in the binary image to determine the object rotation about the mean of the object pixels. Existing PCA-based techniques align the data in their principal spread; however, they have a critical problem of 180° rotation in their principal axes. The current paper provides an automatic solution to address this problem. The algorithm is based on assignment rules inferred from the eigenvectors given by PCA. We applied the proposed algorithm to different datasets including handwritten images of digits, a rotated fingerprint image dataset, and a dataset of magnetic resonance brain images, and confirmed that the proposed method aligns the data efficiently and accurately. In addition to alignment, the algorithm proposes two standard orientations for automatically assessing the true side (upside) of an object.

7 November 2023 in Expert Systems with Applications
Verified
Journal Article
Open access
Redhwan AlgabriRedhwan Algabri
Redhwan Algabri
HSHyunsoo Shin
Hyunsoo Shin
Sungon LeeSungon Lee
Sungon Lee

25 June 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
WRWooseok Ryu
Wooseok Ryu
Youngjin ChoiYoungjin Choi
Youngjin Choi
YCYong Je Choi
Yong Je Choi
YLYeong Geol Lee
Yeong Geol Lee
Sungon LeeSungon Lee
Sungon Lee
An anthropomorphic prosthetic hand for wrist or forearm amputees is developed herein. The prosthetic hand was designed with an underactuated mechanism, which makes self-adaptive grasping possible, as well as natural motions such as flexion and extension. The finger and thumb modules were designed with four degrees of freedom by motions of the distal interphalangeal, proximal interphalangeal, and metacarpophalangeal joints. In this research, we pursued several novel trials in prosthetic hand design. By using two four-bar linkages composed of a combination of linkages and gears for coupling joints at each finger, it was possible to make a compact design, and the linkage has advantages such as accurate positioning, uniform power transmission, and high payload. Also, by using constant-velocity joints, torque is transferred to finger modules regardless of adduction/abduction motions. In addition, adduction/abduction and self-adaptive grasping motions are passively realized using torsional springs. The developed prosthetic hand was fabricated with a weight of 475 g and a human hand size of 175 mm. Experiments with diverse objects showed its good functionality.

20 May 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
WRWooseok Ryu
Wooseok Ryu
Youngjin ChoiYoungjin Choi
Youngjin Choi
YCYong Je Choi
Yong Je Choi
Sungon LeeSungon Lee
Sungon Lee
Finger amputations are the most common upper limb amputation, and they occur approximately 100 times more often than hand amputations. We developed a prosthetic hand for amputees with a thumb and metacarpal. In this case of amputation, the installation of actuators and electrical components is difficult because space is considerably limited owing to the residual metacarpal. This design issue is solved by installing actuators vertically between finger modules and the mounting part where the electrical parts are embedded. With this design, the arrangement between the palm of the patient and the fingers of the prosthetic hand can be adjusted as well. Weight is also an especially important design issue in the development of a practical prosthetic hand because the patient perceives that weight. Most prosthetic hands are attached to the residual limb of the amputee by compressing the contact surface between the product and the amputated body part of the patient. Heavy weight causes users to feel discomfort and fatigue over extended periods of usage. In this study, all frames used in the proposed prosthetic hand were fabricated from nylon using multi-jet-fusion three-dimensional printing. As a result, the weight of the developed prosthetic hand was only 152.32 g but still had the desired strength and stiffness. Each prosthetic finger has four-degree-of-freedom. The distal interphalangeal, proximal interphalangeal, and metacarpophalangeal joints are coupled and driven by power from an actuator, which is transferred to each joint through the tendons. Adduction, abduction, and self-adaptive grasping motions were passively realized using linear and torsional springs. The prototype was fabricated based on these design concepts and functions, and its functionality was verified in experiments using diverse objects.