Sangui Choi

Dr. Sangui Choi

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
16 August 2024 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Verified
Journal Article
Open access
KSKwangsub Song
Kwangsub Song
HSHyung Eun Shin
Hyung Eun Shin
WPWookhyun Park
Wookhyun Park
DLDaehyun Lee
Daehyun Lee
JJJaeyoung Jang
Jaeyoung Jang
CWChang Won Won
Chang Won Won
Sarcopenia is a comprehensive degenerative disease with the progressive loss of skeletal muscle mass with age, accompanied by the loss of muscle strength and muscle dysfunction. Individuals with unmanaged sarcopenia may experience adverse outcomes. Periodically monitoring muscle function to detect muscle degeneration caused by sarcopenia and treating degenerated muscles is essential. We proposed a digital biomarker measurement technique using surface electromyography (sEMG) with electrical stimulation and wearable device to conveniently monitor muscle function at home. When motor neurons and muscle fibers are electrically stimulated, stimulated muscle contraction signals (SMCSs) can be obtained using an sEMG sensor. As motor neuron activation is important for muscle contraction and strength, their action potentials for electrical stimulation represent the muscle function. Thus, the SMCSs are closely related to muscle function, presumptively. Using the SMCSs data, a feature vector concatenating spectrogram-based features and deep learning features extracted from a convolutional neural network model using continuous wavelet transform images was used as the input to train a regression model for measuring the digital biomarker. To verify muscle function measurement technique, we recruited 98 healthy participants aged 20–60 years including 48 [49%] men who volunteered for this study. The Pearson correlation coefficient between the label and model estimates was 0.89, suggesting that the proposed model can robustly estimate the label using SMCSs, with mean error and standard deviation of -0.06 and 0.68, respectively. In conclusion, measuring muscle function using the proposed system that involves SMCSs is feasible.

1 March 2023 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Verified
Journal Article
Open access
KSKwangsub Song
Kwangsub Song
SCSangui Choi
Sangui Choi
JSJin-Hong Shin
Jin-Hong Shin
WSWoo-Hyeon Son
Woo-Hyeon Son
HPHyuntae Park
Hyuntae Park
NPNam-Jong Paik
Nam-Jong Paik
We propose a digital biomarker related to muscle strength and muscle endurance (DB/MS and DB/ME) for the diagnosis of muscle disorders based on a multi-layer perceptron (MLP) using stimulated muscle contraction. When muscle mass is reduced in patients with muscle-related diseases or disorders, measurement of DBs that are related to muscle strength and endurance is needed to suitably recover damaged muscles through rehabilitation training. Furthermore, it is difficult to measure DBs using traditional methods at home without an expert; moreover, the measuring equipment is expensive. Additionally, because traditional measurements depend on the subject’s volition, we propose a DB measurement technique that is unaffected by the subject’s volition. To achieve this, we employed an impact response signal (IRS) based on multi-frequency electrical stimulation (MFES) using an electromyography sensor. The feature vector was then extracted using the signal. Because the IRS is obtained from stimulated muscle contraction, which is caused by electrical stimulation, it provides biomedical information about the muscle. Finally, to estimate the strength and endurance of the muscle, the feature vector was passed through the DB estimation model learned through the MLP. To evaluate the performance of the DB measurement algorithm, we collected the MFES-based IRS database for 50 subjects and tested the model with quantitative evaluation methods using the reference for the DB. The reference was measured using torque equipment. The results were compared with the reference, indicating that it is possible to check for muscle disorders which cause decreased physical performance using the proposed algorithm.

1 March 2024 in Life
Verified
Journal Article
Open access
PDF file
JYJae Wan Yoo
Jae Wan Yoo
GPGeun-Young Park
Geun-Young Park
HPHae-Yeon Park
Hae-Yeon Park
YYYeun Jie Yoo
Yeun Jie Yoo
MYMi-Jeong Yoon
Mi-Jeong Yoon
Seong Hoon LimSeong Hoon Lim
Seong Hoon Lim
Background: We aimed to develop a consensus on the need for and priorities of exercise to treat preexisting sarcopenia with hemiplegic stroke. Methods: A modified three-round Delphi study was conducted. The panelists responded to the questionnaire on a 7-point Likert scale. Responses were returned with descriptive statistics in the next round. Consensus was defined as >75% agreement (score of 5–7) with a median > 5. The percentage of strong agreement (score of 6–7) and Kendall’s coefficient of concordance were calculated to demonstrate a more refined interpretation of the consensus. Results: Fifteen panelists contributed to all rounds. The need for exercise was demonstrated. The consensus was reached on 53 of 58 items in the first round and all items in the second and final rounds. The percentage of strong agreement was high for all but eight items. Conclusions: This study is the first Delphi study to investigate the need for and priorities of exercise for treating preexisting sarcopenia in stroke hemiplegia. We present a standard recommendation including 57 priorities and a strong recommendation including 49 priorities. The eight items that were excluded reflected factors that are less important to hemiplegic patients with poor balance, cognitive decline, or mental vulnerability.

1 January 2024 in Brain & Neurorehabilitation
Verified
Journal Article
Open access
PDF file
YJYerim Ji
Yerim Ji
MYMi-Jeong Yoon
Mi-Jeong Yoon
KSKwangsub Song
Kwangsub Song
SCSangui Choi
Sangui Choi
HLHooman Lee
Hooman Lee
SISun Im
Sun Im
Ji Y, et al. Brain Neurorehabil. 2024 May;17(2):e10. https://doi.org/10.12786/bn.2024.17.e10

1 January 2024 in Brain & Neurorehabilitation
Verified
Journal Article
Open access
PDF file
KSKwangsub Song
Kwangsub Song
HPHae-Yeon Park
Hae-Yeon Park
SCSangui Choi
Sangui Choi
SSSeungyup Song
Seungyup Song
HRHanee Rim
Hanee Rim
SISun Im
Sun Im
Song K, et al. Brain Neurorehabil. 2024 Jun;17(2):e12. https://doi.org/10.12786/bn.2024.17.e12

17 September 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
KSKwangsub Song
Kwangsub Song
SCSangui Choi
Sangui Choi
HLHooman Lee
Hooman Lee
In this paper, we propose the long–short-term memory (LSTM)-based voluntary and non-voluntary (VNV) muscle contraction classification algorithm in an electrical stimulation (ES) environment. In order to measure the muscle quality (MQ), we employ the non-voluntary muscle contraction signal, which occurs by the ES. However, if patient movement, such as voluntary muscle contractionm, occurs during the ES, the electromyography (EMG) sensor captures the VNV muscle contraction signals. In addition, the voluntary muscle contraction signal is a noise component in the MQ measurement technique, which uses only non-voluntary muscle contraction signals. For this reason, we need the VNV muscle contraction classification algorithm to classify the mixed EMG signal. In addition, when recording EMG while using the ES, the EMG signal is significantly contaminated due to the ES signal. Therefore, after we suppress the artifact noise, which is contained in the EMG signal, we perform VNV muscle contraction classification. For this, we first eliminate the artifact noise signal using the ES suppression algorithm. Then, we extract the feature vector, and then the feature vector is reconstructed through the feature selection process. Finally, we design the LSTM-based classification model and compare the proposed algorithm with the conventional method using the EMG data. In addition, to verify the performance of the proposed algorithm, we quantitatively compared results in terms of the confusion matrix and total accuracy. As a result, the performance of the proposed algorithm was higher than that of the conventional methods, including the support vector machine (SVM), artificial neural network (ANN), and deep neural network (DNN).