Yeonkyeong Kim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yeonkyeong Kim has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Kyuseok Kim
Institute of Human Convergence Health Science, Gachon University, 191, Hambangmoe-ro, Yeonsu-gu, Incheon 21936, Republic of Korea
3 shared publications - Su Hwan Lee
Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul 03722, Republic of Korea
3 shared publications - Kyu Bom Kim
TS Corporation, 211, Hwarang-ro, Seongbuk-gu, Seoul 02772, Republic of Korea
2 shared publications - Ah Young Leem
Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul 03722, Republic of Korea
1 shared publication Kyu Bom Kim
Yonsei Univ, Coll Hlth Sci, Dept Radiat Convergence Engn, Wonju, South Korea
1 shared publication- Young Sam Kim
Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul 03722, Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 1 August 2025
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 14 (15) , 5437
Enhanced Respiratory Sound Classification Using Deep Learning and Multi-Channel Auscultation
Background/Objectives: Identifying and classifying abnormal lung sounds is essential for diagnosing patients with respiratory disorders. In particular, the simultaneous recording of auscultation signals from multiple clinically relevant positions offers greater diagnostic potential compared to traditional single-channel measurements. This study aims to improve the accuracy of respiratory sound classification by leveraging multichannel signals and capturing positional characteristics from multiple sites in the same patient. Methods: We evaluated the performance of respiratory sound classification using multichannel lung sound data with a deep learning model that combines a convolutional neural network (CNN) and long short-term memory (LSTM), based on mel-frequency cepstral coefficients (MFCCs). We analyzed the impact of the number and placement of channels on classification performance. Results: The results demonstrated that using four-channel recordings improved accuracy, sensitivity, specificity, precision, and F1-score by approximately 1.11, 1.15, 1.05, 1.08, and 1.13 times, respectively, compared to using three, two, or single-channel recordings. Conclusions: This study confirms that multichannel data capture a richer set of features corresponding to various respiratory sound characteristics, leading to significantly improved classification performance. The proposed method holds promise for enhancing sound classification accuracy not only in clinical applications but also in broader domains such as speech and audio processing.
- 1 August 2025
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 14 (15) , 5437
Open AccessEnhanced Respiratory Sound Classification Using Deep Learning and Multi-Channel Auscultation
Background/Objectives: Identifying and classifying abnormal lung sounds is essential for diagnosing patients with respiratory disorders. In particular, the simultaneous recording of auscultation signals from multiple clinically relevant positions offers greater diagnostic potential compared to traditional single-channel measurements. This study aims to improve the accuracy of respiratory sound classification by leveraging multichannel signals and capturing positional characteristics from multiple sites in the same patient. Methods: We evaluated the performance of respiratory sound classification using multichannel lung sound data with a deep learning model that combines a convolutional neural network (CNN) and long short-term memory (LSTM), based on mel-frequency cepstral coefficients (MFCCs). We analyzed the impact of the number and placement of channels on classification performance. Results: The results demonstrated that using four-channel recordings improved accuracy, sensitivity, specificity, precision, and F1-score by approximately 1.11, 1.15, 1.05, 1.08, and 1.13 times, respectively, compared to using three, two, or single-channel recordings. Conclusions: This study confirms that multichannel data capture a richer set of features corresponding to various respiratory sound characteristics, leading to significantly improved classification performance. The proposed method holds promise for enhancing sound classification accuracy not only in clinical applications but also in broader domains such as speech and audio processing.
- Open Access
- 5 August 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (15) , 6842
New Web-Based Ventilator Monitoring System Consisting of Central and Remote Mobile Applications in Intensive Care Units
A ventilator central monitoring system (VCMS) that can efficiently respond to and treat patients’ respiratory issues in intensive care units (ICUs) is critical. Using Internet of Things (IoT) technology without loss or delay in patient monitoring data, clinical staff can overcome spatial constraints in patient respiratory management by integrated monitoring of multiple ventilators and providing real-time information through remote mobile applications. This study aimed to establish a VCMS and assess its effectiveness in an ICU setting. A VCMS comprises central monitoring and mobile applications, with significant real-time information from multiple patient monitors and ventilator devices stored and managed through the VCMS server, establishing an integrated monitoring environment on a web-based platform. The developed VCMS was analyzed in terms of real-time display and data transmission. Twenty-one respiratory physicians and staff members participated in usability and satisfaction surveys on the developed VCMS. The data transfer capacity derived an error of approximately , and the difference in data transmission capacity was approximately with a 95% confidence interval of to among 18 ventilators and patient monitors. The proposed VCMS could transmit data from various devices without loss of information within the ICU. The medical software validation, consisting of 37 tasks and 9 scenarios, showed a task completion rate of approximately 92%, with a 95% confidence interval of 88.81–90.43. The satisfaction survey consisted of 23 items and showed results of approximately 4.66 points out of 5. These results demonstrated that the VCMS can be readily used by clinical ICU staff, confirming its clinical utility and applicability. The proposed VCMS can help clinical staff quickly respond to the alarm of abnormal events and diagnose and treat based on longitudinal patient data. The mobile applications overcame space constraints, such as isolation to prevent respiratory infection transmission of clinical staff for continuous monitoring of respiratory patients and enabled rapid consultation, ensuring consistent care.- 5 August 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (15) , 6842
Open AccessNew Web-Based Ventilator Monitoring System Consisting of Central and Remote Mobile Applications in Intensive Care Units
A ventilator central monitoring system (VCMS) that can efficiently respond to and treat patients’ respiratory issues in intensive care units (ICUs) is critical. Using Internet of Things (IoT) technology without loss or delay in patient monitoring data, clinical staff can overcome spatial constraints in patient respiratory management by integrated monitoring of multiple ventilators and providing real-time information through remote mobile applications. This study aimed to establish a VCMS and assess its effectiveness in an ICU setting. A VCMS comprises central monitoring and mobile applications, with significant real-time information from multiple patient monitors and ventilator devices stored and managed through the VCMS server, establishing an integrated monitoring environment on a web-based platform. The developed VCMS was analyzed in terms of real-time display and data transmission. Twenty-one respiratory physicians and staff members participated in usability and satisfaction surveys on the developed VCMS. The data transfer capacity derived an error of approximately , and the difference in data transmission capacity was approximately with a 95% confidence interval of to among 18 ventilators and patient monitors. The proposed VCMS could transmit data from various devices without loss of information within the ICU. The medical software validation, consisting of 37 tasks and 9 scenarios, showed a task completion rate of approximately 92%, with a 95% confidence interval of 88.81–90.43. The satisfaction survey consisted of 23 items and showed results of approximately 4.66 points out of 5. These results demonstrated that the VCMS can be readily used by clinical ICU staff, confirming its clinical utility and applicability. The proposed VCMS can help clinical staff quickly respond to the alarm of abnormal events and diagnose and treat based on longitudinal patient data. The mobile applications overcame space constraints, such as isolation to prevent respiratory infection transmission of clinical staff for continuous monitoring of respiratory patients and enabled rapid consultation, ensuring consistent care. - Open Access
- 25 May 2024
- Published by Elsevier in Nuclear Engineering and Technology
- Vol. 56 (10) , 4127-4133
Noise reduction in low-dose positron emission tomography with adaptive parameter estimation in sinogram domain
- 25 May 2024
- Published by Elsevier in Nuclear Engineering and Technology
- Vol. 56 (10) , 4127-4133
Open AccessNoise reduction in low-dose positron emission tomography with adaptive parameter estimation in sinogram domain