Jimin Lee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jimin Lee has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Young-Eun An
IT Research Institute, Chosun University,Gwangju,Korea
3 shared publications - Eashita Wazed
Department of Artificial Intelligence Convergence, Chonnam National University, Gwangju 61186, Republic of Korea
2 shared publications - Hieyong Jeong
Department of Artificial Intelligence Convergence, Chonnam National University, Gwangju, Republic of Korea
2 shared publications - Jong-An Park
The College of Electronic and Information Engineering, Chosun University, Gwangju, South Korea
2 shared publications - Jin-Ho You1 shared publication
- Jong-Hun Chun1 shared publication
Latest Publications
- Open Access
Deep Learning for Heart Sound Abnormality of Infants: Proof-of-Concept Study of 1D and 2D Representations
Introduction: Advanced identification and intervention for Congenital Heart Defects (CHDs) in pediatric populations are crucial, as approximately 1% of neonates worldwide present with these conditions. Traditional methods of diagnosing CHDs often rely on stethoscope auscultation, which heavily depends on the clinician’s expertise and may lead to the oversight of subtle acoustic indicators. Objectives: This study introduces an innovative deep-learning framework designed for the early diagnosis of congenital heart disease. It utilizes time-series data obtained from cardiac auditory signals captured through stethoscopes. Methods: The audio signals were processed into time–frequency representations using Mel-Frequency Cepstral Coefficients (MFCCs). The architecture of the model combines Convolutional Neural Networks (CNNs) for effective feature extraction with Long Short-Term Memory (LSTM) networks to accurately model temporal dependencies. Impressively, the model achieved an accuracy of 98.91% in the early detection of CHDs. Results: While traditional diagnostic tools like Electrocardiograms (ECG) and Phonocardiograms (PCG) remain indispensable for confirming diagnoses, many AI studies have primarily targeted ECG and PCG datasets. This approach emphasizes the potential of cardiac acoustics for the early diagnosis of CHDs, which could lead to improved clinical outcomes for infants. Notably, the dataset used in this research is publicly available, enabling wider application and model training within the research community.
Open AccessDeep Learning for Heart Sound Abnormality of Infants: Proof-of-Concept Study of 1D and 2D Representations
Introduction: Advanced identification and intervention for Congenital Heart Defects (CHDs) in pediatric populations are crucial, as approximately 1% of neonates worldwide present with these conditions. Traditional methods of diagnosing CHDs often rely on stethoscope auscultation, which heavily depends on the clinician’s expertise and may lead to the oversight of subtle acoustic indicators. Objectives: This study introduces an innovative deep-learning framework designed for the early diagnosis of congenital heart disease. It utilizes time-series data obtained from cardiac auditory signals captured through stethoscopes. Methods: The audio signals were processed into time–frequency representations using Mel-Frequency Cepstral Coefficients (MFCCs). The architecture of the model combines Convolutional Neural Networks (CNNs) for effective feature extraction with Long Short-Term Memory (LSTM) networks to accurately model temporal dependencies. Impressively, the model achieved an accuracy of 98.91% in the early detection of CHDs. Results: While traditional diagnostic tools like Electrocardiograms (ECG) and Phonocardiograms (PCG) remain indispensable for confirming diagnoses, many AI studies have primarily targeted ECG and PCG datasets. This approach emphasizes the potential of cardiac acoustics for the early diagnosis of CHDs, which could lead to improved clinical outcomes for infants. Notably, the dataset used in this research is publicly available, enabling wider application and model training within the research community.
- Open Access
- 7 August 2025
- Published by MDPI AG in Preprints.org
Deep Learning for Heart Sound Abnormality of Infants: Proof of Conceptual Study of 1D and 2D Representations
Timely diagnosis and treatment of Congenital Heart Defects (CHDs) in pediatric patients are critical, as approximately 1% of neonates worldwide are affected by these anomalies. Traditional stethoscope auscultation relies on the clinician’s skills, which can lead to missed subtle symptoms. This study introduces a deep-learning framework for the early diagnosis of congenital heart disease, utilizing time-series data from cardiac auditory signals captured via stethoscopes. The audio signals were transformed using Mel Frequency Cepstral Coefficients (MFCC) to create time-frequency representations. Our novel architecture combines Convolutional Neural Networks (CNN) for feature extraction with Long Short-Term Memory (LSTM) networks to capture temporal dependencies. This model achieved an impressive accuracy of 98.91% in early disease detection. While methods such as Electrocardiograms (ECG) and Phonocardiograms (PCG) are necessary for confirming diagnoses, previous AI-driven studies have largely focused on ECG and PCG datasets. Our approach emphasizes the potential of using cardiac acoustics for the early diagnosis of CHDs, enhancing clinical outcomes for infants.
- 7 August 2025
- Published by MDPI AG in Preprints.org
Open AccessDeep Learning for Heart Sound Abnormality of Infants: Proof of Conceptual Study of 1D and 2D Representations
Timely diagnosis and treatment of Congenital Heart Defects (CHDs) in pediatric patients are critical, as approximately 1% of neonates worldwide are affected by these anomalies. Traditional stethoscope auscultation relies on the clinician’s skills, which can lead to missed subtle symptoms. This study introduces a deep-learning framework for the early diagnosis of congenital heart disease, utilizing time-series data from cardiac auditory signals captured via stethoscopes. The audio signals were transformed using Mel Frequency Cepstral Coefficients (MFCC) to create time-frequency representations. Our novel architecture combines Convolutional Neural Networks (CNN) for feature extraction with Long Short-Term Memory (LSTM) networks to capture temporal dependencies. This model achieved an impressive accuracy of 98.91% in early disease detection. While methods such as Electrocardiograms (ECG) and Phonocardiograms (PCG) are necessary for confirming diagnoses, previous AI-driven studies have largely focused on ECG and PCG datasets. Our approach emphasizes the potential of using cardiac acoustics for the early diagnosis of CHDs, enhancing clinical outcomes for infants.
- 28 February 2022
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 20 (2) , 29-36
Usability Evaluation of Thermal Image Database for Emotion Recognition based on Facial Expression Classification
얼굴 표정은 인간 의사소통의 중요한 부분으로 다른 사람의 의도를 이해하는 데 도움이 되는 요소이다. 최근에는 열화상이 감정 인식 분야에서 보완적인 솔루션으로서 가시광선 영상의 단점을 보완하는 대체 수단으로 관심이 높아지고 있다. 본 논문에서는 열화상 데이터를 자체적으로 취득하여 감정 인식에 필요한 얼굴 영역만 따로 추출한 데이터베이스를 구축하고, 구축한 데이터베이스가 감정 인식을 위한 표정 분류에 활용 가능한지 인식 성능을 확인하기 위해 기존 CNN 아키텍처를 이용하여 검증정확도와 학습시간을 분석하였다. CNN 네트워크를 통한 분석 결과, ResNet-18에서 최대 81.28%의 검증정확도를 나타내었고, 평균적으로 68.13%의 검증정확도를 나타내었다. 이를 통해 자체 구축한 열화상 DB가 감정 인식 연구에 유용한 것으로 확인되었다.
- 28 February 2022
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 20 (2) , 29-36
Usability Evaluation of Thermal Image Database for Emotion Recognition based on Facial Expression Classification
얼굴 표정은 인간 의사소통의 중요한 부분으로 다른 사람의 의도를 이해하는 데 도움이 되는 요소이다. 최근에는 열화상이 감정 인식 분야에서 보완적인 솔루션으로서 가시광선 영상의 단점을 보완하는 대체 수단으로 관심이 높아지고 있다. 본 논문에서는 열화상 데이터를 자체적으로 취득하여 감정 인식에 필요한 얼굴 영역만 따로 추출한 데이터베이스를 구축하고, 구축한 데이터베이스가 감정 인식을 위한 표정 분류에 활용 가능한지 인식 성능을 확인하기 위해 기존 CNN 아키텍처를 이용하여 검증정확도와 학습시간을 분석하였다. CNN 네트워크를 통한 분석 결과, ResNet-18에서 최대 81.28%의 검증정확도를 나타내었고, 평균적으로 68.13%의 검증정확도를 나타내었다. 이를 통해 자체 구축한 열화상 DB가 감정 인식 연구에 유용한 것으로 확인되었다.
- 31 July 2021
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 19 (7) , 101-112
A Real-time Image Noise Removal Algorithm for Micro-Endoscope System in Nuclear Power Plants
본 논문에서는 원자로 내 고방사선 구역에 설치된 표준형 증기발생기의 상부번들 내부를 검사하기 위한 초소형 내시경 장치를 통해 전송되는 영상 신호에 유입된 영상 잡음을 실시간에 제거할 수 있는 고속 영상 잡음 제거 알고리즘을 제안한다. Total variation(TV) 최소화기반 영상 잡음 제거 기법은 이미지 경계의 블러링이 적으면서 잡음 제거 성능이 뛰어나다고 평가된다. 본 논문에서는 영상 잡음 제거에 효율적인 smoothing norm 함수와 이 함수를 도입한 TV 손실 함수를 제안한다. 심층신경망 학습에 활용되고 있는 여러 최적화 탐색 기법들을 제안한 TV 손실함수 최소화에 적용하여 비교 분석함으로써 Adamax가 수렴 속도가 빠르고 잡음 제거 성능도 우수함을 보인다. 그리고, 제안한 TV기반 잡음제거 알고리즘 Adamax-TV를 엣지 컴퓨터 수준 GPU에서 구현하여 VGA급 원자로 내시경 영상의 잡음 제거를 15~27fps 속도로 실시간으로 처리할 수 있음을 보인다.
- 31 July 2021
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 19 (7) , 101-112
A Real-time Image Noise Removal Algorithm for Micro-Endoscope System in Nuclear Power Plants
본 논문에서는 원자로 내 고방사선 구역에 설치된 표준형 증기발생기의 상부번들 내부를 검사하기 위한 초소형 내시경 장치를 통해 전송되는 영상 신호에 유입된 영상 잡음을 실시간에 제거할 수 있는 고속 영상 잡음 제거 알고리즘을 제안한다. Total variation(TV) 최소화기반 영상 잡음 제거 기법은 이미지 경계의 블러링이 적으면서 잡음 제거 성능이 뛰어나다고 평가된다. 본 논문에서는 영상 잡음 제거에 효율적인 smoothing norm 함수와 이 함수를 도입한 TV 손실 함수를 제안한다. 심층신경망 학습에 활용되고 있는 여러 최적화 탐색 기법들을 제안한 TV 손실함수 최소화에 적용하여 비교 분석함으로써 Adamax가 수렴 속도가 빠르고 잡음 제거 성능도 우수함을 보인다. 그리고, 제안한 TV기반 잡음제거 알고리즘 Adamax-TV를 엣지 컴퓨터 수준 GPU에서 구현하여 VGA급 원자로 내시경 영상의 잡음 제거를 15~27fps 속도로 실시간으로 처리할 수 있음을 보인다.
- 30 September 2020
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 18 (9) , 107-113
Object Retrieval based on Complexity and Frequency Normalization of Contour Chaincode
- 30 September 2020
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 18 (9) , 107-113
Object Retrieval based on Complexity and Frequency Normalization of Contour Chaincode
- 4 June 2016
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 75 (20) , 12983-12997
Image retrieval technique using the clustering based on rearranged radon transform
- 4 June 2016
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 75 (20) , 12983-12997
Image retrieval technique using the clustering based on rearranged radon transform