Gi-Su Shin
Gi-Su Shin has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Dong Geon Lee
Department of Physical Therapy, Sinsegae Rehabilitation Hospital, Changwon 51216, Republic of Korea
1 shared publication - Dongbock Kim
Department of Mathematics Education, Kyungnam University, 7 Kyungnamdaehak-ro, Masanhappo-gu, Changwon-si 51767, Gyeongsangnam-do, Korea
1 shared publication - GyuChang Lee
Department of Physical Therapy, Kyungnam University, Changwon, Republic of Korea
1 shared publication - Ki Hun Cho
Department of Physical Therapy, Korea National University of Transportation, Jeungpyeong 27909, Chungcheongbuk-do, Republic of Korea
1 shared publication - Nam-Hyun Yoo
Department of Computer Engineering, Kyungnam University, Changwon, Republic of Korea
1 shared publication - Sat Byul Seo
Department of Mathematics Education, Kyungnam University, 7 Kyungnamdaehak-ro, Changwon-si 51767, Geongsangnam-do, Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 19 August 2022
- Published by MDPI AG in Computation
- Vol. 10 (8) , 142
A Study on Sample Size Sensitivity of Factory Manufacturing Dataset for CNN-Based Defective Product Classification
In many small- and medium-sized enterprises (SMEs), defective products are still manually verified in the manufacturing process. Recently, image classification applying deep learning technology has been successful in classifying images of defective and intact products, although there are few cases of utilizing it in practice. SMEs have limited resources; therefore, it is crucial to make careful decisions when applying new methods. We investigated sample size sensitivity to determine the stable performance of deep learning models when applied to the real world. A simple sequential model was constructed, and the dataset was reconstructed into several sizes. For each case, we observed its statistical indicators, such as accuracy, recall, precision, and F1 score, on the same test dataset. Additionally, the loss, accuracy, and AUROC values for the validation dataset were investigated during training. As a result of the conducted research, we were able to confirm that, with 1000 data points or more, the accuracy exceeded 97%. However, more than 5000 cases were required to achieve stability in the model, which had little possibility of overfitting.
- 19 August 2022
- Published by MDPI AG in Computation
- Vol. 10 (8) , 142
Open AccessA Study on Sample Size Sensitivity of Factory Manufacturing Dataset for CNN-Based Defective Product Classification
In many small- and medium-sized enterprises (SMEs), defective products are still manually verified in the manufacturing process. Recently, image classification applying deep learning technology has been successful in classifying images of defective and intact products, although there are few cases of utilizing it in practice. SMEs have limited resources; therefore, it is crucial to make careful decisions when applying new methods. We investigated sample size sensitivity to determine the stable performance of deep learning models when applied to the real world. A simple sequential model was constructed, and the dataset was reconstructed into several sizes. For each case, we observed its statistical indicators, such as accuracy, recall, precision, and F1 score, on the same test dataset. Additionally, the loss, accuracy, and AUROC values for the validation dataset were investigated during training. As a result of the conducted research, we were able to confirm that, with 1000 data points or more, the accuracy exceeded 97%. However, more than 5000 cases were required to achieve stability in the model, which had little possibility of overfitting.
- Open Access
- 30 January 2018
- Published by Springer Nature in BioMedical Engineering OnLine
- Vol. 17 (1) , 14
A novel one arm motorized walker for hemiplegic stroke survivors: a feasibility study
A hemiplegic stroke survivor with a moderate to severe gait disturbance may have difficulty walking using a one-arm walker. This study aimed to test the safety and feasibility of a prototype one-arm motorized walker that uses a power-driven device to provide gait assistance to hemiplegic stroke survivors with moderate to severe gait disturbances. A one-arm motorized walker with a power-driven device was developed and tested with respect to 10 distinct variables, including weight, degrees of freedom, handle, handle substitution function, two-sided use function, variable handle height, redirecting function, electric moving parts through the handle control, brake function using the handle control, folding chairs, and design stability. Its safety and feasibility were tested in 19 hemiplegic stroke individuals using the Likert scale and a simple interview. The walker consists of a frame platform including a handle, electric motor for driving, one wheel for driving, two wheels for turning, unlocking sensor, driving button, and turning buttons. The walker is programmed so that a touch sensor in the handle can unlock the locking system. Furthermore, it is programmed so that a user can propel it by pushing the handle downward or pressing a button and can control directions for turning right or left by pressing buttons. Safety and performance testing was achieved for 10 separate variables, and a Likert scale score of 3.5 of 5 was recorded. This walker’s novel design was developed for hemiplegic stroke survivors with moderate to severe gait disturbances. Our findings indicate that the walker is both safe and feasible for providing walking assistance to hemiplegic stroke survivors and establish the potential advantages of the one-arm motorized walker.
- 30 January 2018
- Published by Springer Nature in BioMedical Engineering OnLine
- Vol. 17 (1) , 14
Open AccessA novel one arm motorized walker for hemiplegic stroke survivors: a feasibility study
A hemiplegic stroke survivor with a moderate to severe gait disturbance may have difficulty walking using a one-arm walker. This study aimed to test the safety and feasibility of a prototype one-arm motorized walker that uses a power-driven device to provide gait assistance to hemiplegic stroke survivors with moderate to severe gait disturbances. A one-arm motorized walker with a power-driven device was developed and tested with respect to 10 distinct variables, including weight, degrees of freedom, handle, handle substitution function, two-sided use function, variable handle height, redirecting function, electric moving parts through the handle control, brake function using the handle control, folding chairs, and design stability. Its safety and feasibility were tested in 19 hemiplegic stroke individuals using the Likert scale and a simple interview. The walker consists of a frame platform including a handle, electric motor for driving, one wheel for driving, two wheels for turning, unlocking sensor, driving button, and turning buttons. The walker is programmed so that a touch sensor in the handle can unlock the locking system. Furthermore, it is programmed so that a user can propel it by pushing the handle downward or pressing a button and can control directions for turning right or left by pressing buttons. Safety and performance testing was achieved for 10 separate variables, and a Likert scale score of 3.5 of 5 was recorded. This walker’s novel design was developed for hemiplegic stroke survivors with moderate to severe gait disturbances. Our findings indicate that the walker is both safe and feasible for providing walking assistance to hemiplegic stroke survivors and establish the potential advantages of the one-arm motorized walker.