Chieh-Chun Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chieh-Chun Huang has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
Tzu-Yu Chu
Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan
3 shared publications- Chen-Ching Lin
Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan
2 shared publications Chern-En Chiang
School of Medicine, National Yang Ming Chiao Tung University College of Medicine, Taipei, Taiwan
2 shared publications- Hao-Min Cheng
Department of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan
2 shared publications Shih-Hsien Sung
Division of Cardiology, Department of Internal Medicine, Taipei Veterans General Hospital, Taipei, Taiwan
2 shared publicationsWei-Ting Wang
Division of Cardiology, Department of Internal Medicine, Taipei Veterans General Hospital, Taipei, Taiwan
2 shared publications
Latest Publications
- Open Access
- 3 January 2025
- Published by Springer Nature in npj Biofilms and Microbiomes
- Vol. 11 (1) , 1-12
PreLect: Prevalence leveraged consistent feature selection decodes microbial signatures across cohorts
The intricate nature of microbiota sequencing data—high dimensionality and sparsity—presents a challenge in identifying informative and reproducible microbial features for both research and clinical applications. Addressing this, we introduce PreLect, an innovative feature selection framework that harnesses microbes’ prevalence to facilitate consistent selection in sparse microbiota data. Upon rigorous benchmarking against established feature selection methodologies across 42 microbiome datasets, PreLect demonstrated superior classification capabilities compared to statistical methods and outperformed machine learning-based methods by selecting features with greater prevalence and abundance. A significant strength of PreLect lies in its ability to reliably identify reproducible microbial features across varied cohorts. Applied to colorectal cancer, PreLect identifies key microbes and highlights crucial pathways, such as lipopolysaccharide and glycerophospholipid biosynthesis, in cancer progression. This case study exemplifies PreLect’s utility in discerning clinically relevant microbial signatures. In summary, PreLect’s accuracy and robustness make it a significant advancement in the analysis of complex microbiota data.
- 3 January 2025
- Published by Springer Nature in npj Biofilms and Microbiomes
- Vol. 11 (1) , 1-12
Open AccessPreLect: Prevalence leveraged consistent feature selection decodes microbial signatures across cohorts
The intricate nature of microbiota sequencing data—high dimensionality and sparsity—presents a challenge in identifying informative and reproducible microbial features for both research and clinical applications. Addressing this, we introduce PreLect, an innovative feature selection framework that harnesses microbes’ prevalence to facilitate consistent selection in sparse microbiota data. Upon rigorous benchmarking against established feature selection methodologies across 42 microbiome datasets, PreLect demonstrated superior classification capabilities compared to statistical methods and outperformed machine learning-based methods by selecting features with greater prevalence and abundance. A significant strength of PreLect lies in its ability to reliably identify reproducible microbial features across varied cohorts. Applied to colorectal cancer, PreLect identifies key microbes and highlights crucial pathways, such as lipopolysaccharide and glycerophospholipid biosynthesis, in cancer progression. This case study exemplifies PreLect’s utility in discerning clinically relevant microbial signatures. In summary, PreLect’s accuracy and robustness make it a significant advancement in the analysis of complex microbiota data.
- Open Access
- 15 February 2024
- Published by Springer Nature in Physical and Engineering Sciences in Medicine
- Vol. 47 (2) , 477-489
Examining arterial pulsation to identify and risk-stratify heart failure subjects with deep neural network
Hemodynamic parameters derived from pulse wave analysis have been shown to predict long-term outcomes in patients with heart failure (HF). Here we aimed to develop a deep-learning based algorithm that incorporates pressure waveforms for the identification and risk stratification of patients with HF. The first study, with a case–control study design to address data imbalance issue, included 431 subjects with HF exhibiting typical symptoms and 1545 control participants with no history of HF (non-HF). Carotid pressure waveforms were obtained from all the participants using applanation tonometry. The HF score, representing the probability of HF, was derived from a one-dimensional deep neural network (DNN) model trained with characteristics of the normalized carotid pressure waveform. In the second study of HF patients, we constructed a Cox regression model with 83 candidate clinical variables along with the HF score to predict the risk of all-cause mortality along with rehospitalization. To identify subjects using the HF score, the sensitivity, specificity, accuracy, F1 score, and area under receiver operating characteristic curve were 0.867, 0.851, 0.874, 0.878, and 0.93, respectively, from the hold-out cross-validation of the DNN, which was better than other machine learning models, including logistic regression, support vector machine, and random forest. With a median follow-up of 5.8 years, the multivariable Cox model using the HF score and other clinical variables outperformed the other HF risk prediction models with concordance index of 0.71, in which only the HF score and five clinical variables were independent significant predictors (p < 0.05), including age, history of percutaneous coronary intervention, concentration of sodium in the emergency room, N-terminal pro-brain natriuretic peptide, and hemoglobin. Our study demonstrated the diagnostic and prognostic utility of arterial waveforms in subjects with HF using a DNN model. Pulse wave contains valuable information that can benefit the clinical care of patients with HF.
- 15 February 2024
- Published by Springer Nature in Physical and Engineering Sciences in Medicine
- Vol. 47 (2) , 477-489
Open AccessExamining arterial pulsation to identify and risk-stratify heart failure subjects with deep neural network
Hemodynamic parameters derived from pulse wave analysis have been shown to predict long-term outcomes in patients with heart failure (HF). Here we aimed to develop a deep-learning based algorithm that incorporates pressure waveforms for the identification and risk stratification of patients with HF. The first study, with a case–control study design to address data imbalance issue, included 431 subjects with HF exhibiting typical symptoms and 1545 control participants with no history of HF (non-HF). Carotid pressure waveforms were obtained from all the participants using applanation tonometry. The HF score, representing the probability of HF, was derived from a one-dimensional deep neural network (DNN) model trained with characteristics of the normalized carotid pressure waveform. In the second study of HF patients, we constructed a Cox regression model with 83 candidate clinical variables along with the HF score to predict the risk of all-cause mortality along with rehospitalization. To identify subjects using the HF score, the sensitivity, specificity, accuracy, F1 score, and area under receiver operating characteristic curve were 0.867, 0.851, 0.874, 0.878, and 0.93, respectively, from the hold-out cross-validation of the DNN, which was better than other machine learning models, including logistic regression, support vector machine, and random forest. With a median follow-up of 5.8 years, the multivariable Cox model using the HF score and other clinical variables outperformed the other HF risk prediction models with concordance index of 0.71, in which only the HF score and five clinical variables were independent significant predictors (p < 0.05), including age, history of percutaneous coronary intervention, concentration of sodium in the emergency room, N-terminal pro-brain natriuretic peptide, and hemoglobin. Our study demonstrated the diagnostic and prognostic utility of arterial waveforms in subjects with HF using a DNN model. Pulse wave contains valuable information that can benefit the clinical care of patients with HF.
- Open Access
- 17 May 2023
- Published by Springer Nature in Research Square
Examining Arterial Pulsation to Identify and Risk-Stratify Heart Failure Subjects with Deep Neural Network
Background Hemodynamic parameters derived from pulse wave analysis have been shown to predict long-term outcomes in patients with heart failure (HF). Here we aimed to develop a deep-learning based algorithm that incorporates pressure waveforms for the identification and risk stratification of patients with HF. Methods The first study, with a case-control study design to address data imbalance issue, included 431 subjects with HF exhibiting typical symptoms and a left ventricular ejection fraction (LVEF) less than 45% and 1545 control participants with no history of HF (non-HF). Carotid pressure waveforms were obtained from all the participants using applanation tonometry. The HF score, representing the probability of HF, was derived from a one-dimensional deep neural network (DNN) model trained with characteristics of the normalized carotid pressure waveform. In the second study of HF patients, we constructed a Cox regression model with 83 candidate clinical variables along with the HF score to predict the risk of all-cause mortality with rehospitalization. Results To identify subjects using the HF score, the sensitivity, specificity, accuracy, F1 score, and area under receiver operating characteristic curve were 0.867, 0.851, 0.874, 0.878, and 0.93, respectively, from the 10-fold cross-validation of the DNN, which was better than other machine learning models, including logistic regression, support vector machine, and random forest. With a median follow-up of 5.8 years, the multivariable Cox model using the HF score and other clinical variables outperformed the other HF risk prediction models with concordance index of 0.71, in which only the HF score and five clinical variables were independent significant predictors (p < 0.05), including age, history of percutaneous coronary intervention, concentration of sodium in the emergency room, N-terminal pro-brain natriuretic peptide, and hemoglobin. Conclusions Our study demonstrated the diagnostic and prognostic utility of arterial waveforms in subjects with HF using a DNN model. Pulse wave contains valuable information that can benefit the clinical care of patients with HF.
- 17 May 2023
- Published by Springer Nature in Research Square
Open AccessExamining Arterial Pulsation to Identify and Risk-Stratify Heart Failure Subjects with Deep Neural Network
Background Hemodynamic parameters derived from pulse wave analysis have been shown to predict long-term outcomes in patients with heart failure (HF). Here we aimed to develop a deep-learning based algorithm that incorporates pressure waveforms for the identification and risk stratification of patients with HF. Methods The first study, with a case-control study design to address data imbalance issue, included 431 subjects with HF exhibiting typical symptoms and a left ventricular ejection fraction (LVEF) less than 45% and 1545 control participants with no history of HF (non-HF). Carotid pressure waveforms were obtained from all the participants using applanation tonometry. The HF score, representing the probability of HF, was derived from a one-dimensional deep neural network (DNN) model trained with characteristics of the normalized carotid pressure waveform. In the second study of HF patients, we constructed a Cox regression model with 83 candidate clinical variables along with the HF score to predict the risk of all-cause mortality with rehospitalization. Results To identify subjects using the HF score, the sensitivity, specificity, accuracy, F1 score, and area under receiver operating characteristic curve were 0.867, 0.851, 0.874, 0.878, and 0.93, respectively, from the 10-fold cross-validation of the DNN, which was better than other machine learning models, including logistic regression, support vector machine, and random forest. With a median follow-up of 5.8 years, the multivariable Cox model using the HF score and other clinical variables outperformed the other HF risk prediction models with concordance index of 0.71, in which only the HF score and five clinical variables were independent significant predictors (p < 0.05), including age, history of percutaneous coronary intervention, concentration of sodium in the emergency room, N-terminal pro-brain natriuretic peptide, and hemoglobin. Conclusions Our study demonstrated the diagnostic and prognostic utility of arterial waveforms in subjects with HF using a DNN model. Pulse wave contains valuable information that can benefit the clinical care of patients with HF.