Thien Vu
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
Thien Vu has a total of 63 co-authors. Here are the top co-authors by number of shared publications:
- Michihiro Araki
Graduate School of Science, Technology and Innovation, Kobe University, Kobe, Japan
8 shared publications - Agustin Martin-Morales
Artificial Intelligence Center for Health and Biomedical Research, National Institutes of Biomedical Innovation, Health and Nutrition, Osaka, Japan
7 shared publications - Mai Inoue
Artificial Intelligence Center for Health and Biomedical Research, National Institutes of Biomedical Innovation, Health and Nutrition, 3-17 Senrioka-shinmachi, Osaka, 566-0002, Japan, 81 8093069457
6 shared publications - Masaki Yamamoto
Artificial Intelligence Center for Health and Biomedical Research, National Institutes of Biomedical Innovation, Health and Nutrition, Osaka, Japan
5 shared publications - Research Dawadi
Artificial Intelligence Center for Health and Biomedical Research, National Institutes of Biomedical Innovation, Health and Nutrition, 3-17 Senrioka-shinmachi, Osaka, 566-0002, Japan, 81 8093069457
5 shared publications - Attayeb Mohsen
Artificial Intelligence Center for Health and Biomedical Research (ArCHER), National Institutes of Biomedical Innovation, Health and Nutrition, Osaka, Japan
4 shared publications
Latest Publications
- Open Access
- 16 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 242-247
Comparing Large Language Models for Food-Microbiome Relation Extraction from Research Papers
- 16 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 242-247
Open AccessComparing Large Language Models for Food-Microbiome Relation Extraction from Research Papers
- Top Cited PapersOpen Access
- 12 May 2025
- Published by JMIR Publications Inc. in JMIR Cardio
- Vol. 9, e68066
Machine Learning Model for Predicting Coronary Heart Disease Risk: Development and Validation Using Insights From a Japanese Population–Based Study
Background: Coronary heart disease (CHD) is a major cause of morbidity and mortality worldwide. Identifying key risk factors is essential for effective risk assessment and prevention. A data-driven approach using machine learning (ML) offers advanced techniques to analyze complex, nonlinear, and high-dimensional datasets, uncovering novel predictors of CHD that go beyond the limitations of traditional models, which rely on predefined variables. Objective: This study aims to evaluate the contribution of various risk factors to CHD, focusing on both established and novel markers using ML techniques. Methods: The study recruited 7672 participants aged 30-84 years from Suita City, Japan, between 1989 and 1999. Over an average of 15 years, participants were monitored for cardiovascular events. A total of 7260 participants and 28 variables were included in the analysis after excluding individuals with missing outcome data and eliminating unnecessary variables. Five ML models—logistic regression, random forest (RF), support vector machine, Extreme Gradient Boosting, and Light Gradient-Boosting Machine—were applied for predicting CHD incidence. Model performance was evaluated using accuracy, sensitivity, specificity, precision, area under the curve, F1-score, calibration curves, observed-to-expected ratios, and decision curve analysis. Additionally, Shapley Additive Explanations (SHAPs) were used to interpret the prediction models and understand the contribution of various risk factors to CHD. Results: Among 7260 participants, 305 (4.2%) were diagnosed with CHD. The RF model demonstrated the highest performance, with an accuracy of 0.73 (95% CI 0.64‐0.80), sensitivity of 0.74 (95% CI 0.62‐0.84), specificity of 0.72 (95% CI 0.61‐0.83), and an area under the curve of 0.73 (95% CI 0.65‐0.80). RF also showed excellent calibration, with predicted probabilities closely aligning with observed outcomes, and provided substantial net benefit across a range of risk thresholds, as demonstrated by decision curve analysis. SHAP analysis elucidated key predictors of CHD, including the intima-media thickness (IMT_cMax) of the common carotid artery, blood pressure, lipid profiles (non–high-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglycerides), and estimated glomerular filtration rate. Novel risk factors identified as significant contributors to CHD risk included lower calcium levels, elevated white blood cell counts, and body fat percentage. Furthermore, a protective effect was observed in women, suggesting the potential necessity for gender-specific risk assessment strategies in future cardiovascular health evaluations. Conclusions: We developed a model to predict CHD using ML and applied SHAP methods for interpretation. This approach highlights the multifactor nature of CHD risk evaluation, aiming to support health care professionals in identifying risk factors and formulating effective prevention strategies.
- 12 May 2025
- Published by JMIR Publications Inc. in JMIR Cardio
- Vol. 9, e68066
Top Cited PapersOpen AccessMachine Learning Model for Predicting Coronary Heart Disease Risk: Development and Validation Using Insights From a Japanese Population–Based Study
Background: Coronary heart disease (CHD) is a major cause of morbidity and mortality worldwide. Identifying key risk factors is essential for effective risk assessment and prevention. A data-driven approach using machine learning (ML) offers advanced techniques to analyze complex, nonlinear, and high-dimensional datasets, uncovering novel predictors of CHD that go beyond the limitations of traditional models, which rely on predefined variables. Objective: This study aims to evaluate the contribution of various risk factors to CHD, focusing on both established and novel markers using ML techniques. Methods: The study recruited 7672 participants aged 30-84 years from Suita City, Japan, between 1989 and 1999. Over an average of 15 years, participants were monitored for cardiovascular events. A total of 7260 participants and 28 variables were included in the analysis after excluding individuals with missing outcome data and eliminating unnecessary variables. Five ML models—logistic regression, random forest (RF), support vector machine, Extreme Gradient Boosting, and Light Gradient-Boosting Machine—were applied for predicting CHD incidence. Model performance was evaluated using accuracy, sensitivity, specificity, precision, area under the curve, F1-score, calibration curves, observed-to-expected ratios, and decision curve analysis. Additionally, Shapley Additive Explanations (SHAPs) were used to interpret the prediction models and understand the contribution of various risk factors to CHD. Results: Among 7260 participants, 305 (4.2%) were diagnosed with CHD. The RF model demonstrated the highest performance, with an accuracy of 0.73 (95% CI 0.64‐0.80), sensitivity of 0.74 (95% CI 0.62‐0.84), specificity of 0.72 (95% CI 0.61‐0.83), and an area under the curve of 0.73 (95% CI 0.65‐0.80). RF also showed excellent calibration, with predicted probabilities closely aligning with observed outcomes, and provided substantial net benefit across a range of risk thresholds, as demonstrated by decision curve analysis. SHAP analysis elucidated key predictors of CHD, including the intima-media thickness (IMT_cMax) of the common carotid artery, blood pressure, lipid profiles (non–high-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglycerides), and estimated glomerular filtration rate. Novel risk factors identified as significant contributors to CHD risk included lower calcium levels, elevated white blood cell counts, and body fat percentage. Furthermore, a protective effect was observed in women, suggesting the potential necessity for gender-specific risk assessment strategies in future cardiovascular health evaluations. Conclusions: We developed a model to predict CHD using ML and applied SHAP methods for interpretation. This approach highlights the multifactor nature of CHD risk evaluation, aiming to support health care professionals in identifying risk factors and formulating effective prevention strategies.
- Open Access
- 25 March 2025
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 4 (1) , e59094
Disease Prediction Using Machine Learning on Smartphone-Based Eye, Skin, and Voice Data: Scoping Review
Background: The application of machine learning methods to data generated by ubiquitous devices like smartphones presents an opportunity to enhance the quality of health care and diagnostics. Smartphones are ideal for gathering data easily, providing quick feedback on diagnoses, and proposing interventions for health improvement. Objective: We reviewed the existing literature to gather studies that have used machine learning models with smartphone-derived data for the prediction and diagnosis of health anomalies. We divided the studies into those that used machine learning models by conducting experiments to retrieve data and predict diseases, and those that used machine learning models on publicly available databases. The details of databases, experiments, and machine learning models are intended to help researchers working in the fields of machine learning and artificial intelligence in the health care domain. Researchers can use the information to design their experiments or determine the databases they could analyze. Methods: A comprehensive search of the PubMed and IEEE Xplore databases was conducted, and an in-house keyword screening method was used to filter the articles based on the content of their titles and abstracts. Subsequently, studies related to the 3 areas of voice, skin, and eye were selected and analyzed based on how data for machine learning models were extracted (ie, the use of publicly available databases or through experiments). The machine learning methods used in each study were also noted. Results: A total of 49 studies were identified as being relevant to the topic of interest, and among these studies, there were 31 different databases and 24 different machine learning methods. Conclusions: The results provide a better understanding of how smartphone data are collected for predicting different diseases and what kinds of machine learning methods are used on these data. Similarly, publicly available databases having smartphone-based data that can be used for the diagnosis of various diseases have been presented. Our screening method could be used or improved in future studies, and our findings could be used as a reference to conduct similar studies, experiments, or statistical analyses.
- 25 March 2025
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 4 (1) , e59094
Open AccessDisease Prediction Using Machine Learning on Smartphone-Based Eye, Skin, and Voice Data: Scoping Review
Background: The application of machine learning methods to data generated by ubiquitous devices like smartphones presents an opportunity to enhance the quality of health care and diagnostics. Smartphones are ideal for gathering data easily, providing quick feedback on diagnoses, and proposing interventions for health improvement. Objective: We reviewed the existing literature to gather studies that have used machine learning models with smartphone-derived data for the prediction and diagnosis of health anomalies. We divided the studies into those that used machine learning models by conducting experiments to retrieve data and predict diseases, and those that used machine learning models on publicly available databases. The details of databases, experiments, and machine learning models are intended to help researchers working in the fields of machine learning and artificial intelligence in the health care domain. Researchers can use the information to design their experiments or determine the databases they could analyze. Methods: A comprehensive search of the PubMed and IEEE Xplore databases was conducted, and an in-house keyword screening method was used to filter the articles based on the content of their titles and abstracts. Subsequently, studies related to the 3 areas of voice, skin, and eye were selected and analyzed based on how data for machine learning models were extracted (ie, the use of publicly available databases or through experiments). The machine learning methods used in each study were also noted. Results: A total of 49 studies were identified as being relevant to the topic of interest, and among these studies, there were 31 different databases and 24 different machine learning methods. Conclusions: The results provide a better understanding of how smartphone data are collected for predicting different diseases and what kinds of machine learning methods are used on these data. Similarly, publicly available databases having smartphone-based data that can be used for the diagnosis of various diseases have been presented. Our screening method could be used or improved in future studies, and our findings could be used as a reference to conduct similar studies, experiments, or statistical analyses.
- Top Cited PapersOpen Access
- 17 February 2025
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 25 (1) , 1-12
Prediction of depressive disorder using machine learning approaches: findings from the NHANES
Depressive disorder, particularly major depressive disorder (MDD), significantly impact individuals and society. Traditional analysis methods often suffer from subjectivity and may not capture complex, non-linear relationships between risk factors. Machine learning (ML) offers a data-driven approach to predict and diagnose depression more accurately by analyzing large and complex datasets. This study utilized data from the National Health and Nutrition Examination Survey (NHANES) 2013–2014 to predict depression using six supervised ML models: Logistic Regression, Random Forest, Naive Bayes, Support Vector Machine (SVM), Extreme Gradient Boost (XGBoost), and Light Gradient Boosting Machine (LightGBM). Depression was assessed using the Patient Health Questionnaire (PHQ-9), with a score of 10 or higher indicating moderate to severe depression. The dataset was split into training and testing sets (80% and 20%, respectively), and model performance was evaluated using accuracy, sensitivity, specificity, precision, AUC, and F1 score. SHAP (SHapley Additive exPlanations) values were used to identify the critical risk factors and interpret the contributions of each feature to the prediction. XGBoost was identified as the best-performing model, achieving the highest accuracy, sensitivity, specificity, precision, AUC, and F1 score. SHAP analysis highlighted the most significant predictors of depression: the ratio family income to poverty (PIR), sex, hypertension, serum cotinine and hydroxycotine, BMI, education level, glucose levels, age, marital status, and renal function (eGFR). We developed ML models to predict depression and utilized SHAP for interpretation. This approach identifies key factors associated with depression, encompassing socioeconomic, demographic, and health-related aspects.
- 17 February 2025
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 25 (1) , 1-12
Top Cited PapersOpen AccessPrediction of depressive disorder using machine learning approaches: findings from the NHANES
Depressive disorder, particularly major depressive disorder (MDD), significantly impact individuals and society. Traditional analysis methods often suffer from subjectivity and may not capture complex, non-linear relationships between risk factors. Machine learning (ML) offers a data-driven approach to predict and diagnose depression more accurately by analyzing large and complex datasets. This study utilized data from the National Health and Nutrition Examination Survey (NHANES) 2013–2014 to predict depression using six supervised ML models: Logistic Regression, Random Forest, Naive Bayes, Support Vector Machine (SVM), Extreme Gradient Boost (XGBoost), and Light Gradient Boosting Machine (LightGBM). Depression was assessed using the Patient Health Questionnaire (PHQ-9), with a score of 10 or higher indicating moderate to severe depression. The dataset was split into training and testing sets (80% and 20%, respectively), and model performance was evaluated using accuracy, sensitivity, specificity, precision, AUC, and F1 score. SHAP (SHapley Additive exPlanations) values were used to identify the critical risk factors and interpret the contributions of each feature to the prediction. XGBoost was identified as the best-performing model, achieving the highest accuracy, sensitivity, specificity, precision, AUC, and F1 score. SHAP analysis highlighted the most significant predictors of depression: the ratio family income to poverty (PIR), sex, hypertension, serum cotinine and hydroxycotine, BMI, education level, glucose levels, age, marital status, and renal function (eGFR). We developed ML models to predict depression and utilized SHAP for interpretation. This approach identifies key factors associated with depression, encompassing socioeconomic, demographic, and health-related aspects.
- Open Access
- 29 October 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Machine Learning-Based Prediction of Coronary Heart Disease: Comprehensive Insights from the Suita Study (Preprint)
BACKGROUND Coronary heart disease (CHD) is a major cause of morbidity and mortality worldwide. Identifying key risk factors is essential for effective risk assessment and prevention. Machine learning (ML) offers advanced methods for analyzing complex datasets, revealing novel predictors of CHD beyond traditional models. OBJECTIVE This study aims to evaluate the contribution of various risk factors to CHD, focusing on both established and novel markers using machine learning techniques. METHODS The study recruited 7,672 participants aged 30 to 84 years from Suita City, Japan, between 1989 and 1999. Over an average of 15 years, participants were monitored for cardiovascular events. Five ML models—Random Forest (RF), XGBoost, Support Vector Machine (SVM), Logistic Regression (LR), and LightGBM—were used. The optimal model was identified based on accuracy, sensitivity, specificity, and AUC. SHapley Additive exPlanations (SHAP) were then employed to explore the contribution of various risk factors to CHD. RESULTS RF achieved the highest AUC (95% CI) of 0.94 (0.93-0.96), outperforming LR, SVM, XGBoost, and LightGBM. SHAP on the best model identified the top CHD predictors. Intima-media thickness of common carotid artery (IMT_cMax) was identified as the strongest predictor of CHD, highlighting the importance of arterial health. Systolic and diastolic blood pressure, along with lipid profiles (non-HDL cholesterol, HDL cholesterol, and triglycerides), were closely associated with CHD incidence. eGFR underscored the link between renal function and CHD. Novel insights included the impact of lower calcium levels, systemic inflammation (elevated WBC counts), fructosamine levels, and obesity-related factor (body fat percentage). A protective effect in females indicated the need for sex-specific CHD management strategies. CONCLUSIONS ML, particularly the RF model combined with SHAP, effectively identified key risk factors for CHD, including arterial health, blood pressure, lipid profiles, renal function, and novel markers. These findings support a multifactorial approach to CHD risk assessment.
- 29 October 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessMachine Learning-Based Prediction of Coronary Heart Disease: Comprehensive Insights from the Suita Study (Preprint)
BACKGROUND Coronary heart disease (CHD) is a major cause of morbidity and mortality worldwide. Identifying key risk factors is essential for effective risk assessment and prevention. Machine learning (ML) offers advanced methods for analyzing complex datasets, revealing novel predictors of CHD beyond traditional models. OBJECTIVE This study aims to evaluate the contribution of various risk factors to CHD, focusing on both established and novel markers using machine learning techniques. METHODS The study recruited 7,672 participants aged 30 to 84 years from Suita City, Japan, between 1989 and 1999. Over an average of 15 years, participants were monitored for cardiovascular events. Five ML models—Random Forest (RF), XGBoost, Support Vector Machine (SVM), Logistic Regression (LR), and LightGBM—were used. The optimal model was identified based on accuracy, sensitivity, specificity, and AUC. SHapley Additive exPlanations (SHAP) were then employed to explore the contribution of various risk factors to CHD. RESULTS RF achieved the highest AUC (95% CI) of 0.94 (0.93-0.96), outperforming LR, SVM, XGBoost, and LightGBM. SHAP on the best model identified the top CHD predictors. Intima-media thickness of common carotid artery (IMT_cMax) was identified as the strongest predictor of CHD, highlighting the importance of arterial health. Systolic and diastolic blood pressure, along with lipid profiles (non-HDL cholesterol, HDL cholesterol, and triglycerides), were closely associated with CHD incidence. eGFR underscored the link between renal function and CHD. Novel insights included the impact of lower calcium levels, systemic inflammation (elevated WBC counts), fructosamine levels, and obesity-related factor (body fat percentage). A protective effect in females indicated the need for sex-specific CHD management strategies. CONCLUSIONS ML, particularly the RF model combined with SHAP, effectively identified key risk factors for CHD, including arterial health, blood pressure, lipid profiles, renal function, and novel markers. These findings support a multifactorial approach to CHD risk assessment.
- Open Access
- 1 July 2024
- Published by MDPI AG in Journal of Cardiovascular Development and Disease
- Vol. 11 (7) , 207
Machine Learning Approaches for Stroke Risk Prediction: Findings from the Suita Study
Stroke constitutes a significant public health concern due to its impact on mortality and morbidity. This study investigates the utility of machine learning algorithms in predicting stroke and identifying key risk factors using data from the Suita study, comprising 7389 participants and 53 variables. Initially, unsupervised k-prototype clustering categorized participants into risk clusters, while five supervised models including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosted Machine (LightGBM) were employed to predict stroke outcomes. Stroke incidence disparities among identified risk clusters using the unsupervised k-prototype clustering method are substantial, according to the findings. Supervised learning, particularly RF, was a preferable option because of the higher levels of performance metrics. The Shapley Additive Explanations (SHAP) method identified age, systolic blood pressure, hypertension, estimated glomerular filtration rate, metabolic syndrome, and blood glucose level as key predictors of stroke, aligning with findings from the unsupervised clustering approach in high-risk groups. Additionally, previously unidentified risk factors such as elbow joint thickness, fructosamine, hemoglobin, and calcium level demonstrate potential for stroke prediction. In conclusion, machine learning facilitated accurate stroke risk predictions and highlighted potential biomarkers, offering a data-driven framework for risk assessment and biomarker discovery.
- 1 July 2024
- Published by MDPI AG in Journal of Cardiovascular Development and Disease
- Vol. 11 (7) , 207
Open AccessMachine Learning Approaches for Stroke Risk Prediction: Findings from the Suita Study
Stroke constitutes a significant public health concern due to its impact on mortality and morbidity. This study investigates the utility of machine learning algorithms in predicting stroke and identifying key risk factors using data from the Suita study, comprising 7389 participants and 53 variables. Initially, unsupervised k-prototype clustering categorized participants into risk clusters, while five supervised models including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosted Machine (LightGBM) were employed to predict stroke outcomes. Stroke incidence disparities among identified risk clusters using the unsupervised k-prototype clustering method are substantial, according to the findings. Supervised learning, particularly RF, was a preferable option because of the higher levels of performance metrics. The Shapley Additive Explanations (SHAP) method identified age, systolic blood pressure, hypertension, estimated glomerular filtration rate, metabolic syndrome, and blood glucose level as key predictors of stroke, aligning with findings from the unsupervised clustering approach in high-risk groups. Additionally, previously unidentified risk factors such as elbow joint thickness, fructosamine, hemoglobin, and calcium level demonstrate potential for stroke prediction. In conclusion, machine learning facilitated accurate stroke risk predictions and highlighted potential biomarkers, offering a data-driven framework for risk assessment and biomarker discovery.
- Open Access
- 14 May 2024
- Published by MDPI AG in Preprints.org
Machine Learning Approaches for Stroke Risk Prediction: Findings from the Suita Study
Stroke constitutes a significant public health concern due to its impact on mortality and morbidity. This study investigates the utility of machine learning algorithms in predicting stroke and identifying key risk factors using data from the Suita study, comprising 7,389 participants and 53 variables. Initially, unsupervised K-prototype clustering categorized participants into risk clusters, while five supervised models including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosted Machine (Light-GBM) were employed to predict stroke outcomes. Stroke incidence disparities among identified risk clusters using the unsupervised K-prototype clustering method are substantial, according to the findings. Supervised learning, particularly RF was a preferable option because of the higher levels of performance metrics. The Shapley Additive Explanations (SHAP) method identified age, systolic blood pressure, hypertension, estimated glomerular filtration rate, metabolic syndrome, and blood glucose level as key predictors of stroke, aligning with findings from the unsupervised clustering approach in high-risk groups. Additionally, previously unidentified risk factors such as elbow joint thickness, fructosamine, hemoglobin, and calcium level demonstrate potential for stroke prediction. In conclusion, machine learning facilitated accurate stroke risk predictions and highlighted potential biomarkers, offering a data-driven framework for risk assessment and biomarker discovery.
- 14 May 2024
- Published by MDPI AG in Preprints.org
Open AccessMachine Learning Approaches for Stroke Risk Prediction: Findings from the Suita Study
Stroke constitutes a significant public health concern due to its impact on mortality and morbidity. This study investigates the utility of machine learning algorithms in predicting stroke and identifying key risk factors using data from the Suita study, comprising 7,389 participants and 53 variables. Initially, unsupervised K-prototype clustering categorized participants into risk clusters, while five supervised models including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosted Machine (Light-GBM) were employed to predict stroke outcomes. Stroke incidence disparities among identified risk clusters using the unsupervised K-prototype clustering method are substantial, according to the findings. Supervised learning, particularly RF was a preferable option because of the higher levels of performance metrics. The Shapley Additive Explanations (SHAP) method identified age, systolic blood pressure, hypertension, estimated glomerular filtration rate, metabolic syndrome, and blood glucose level as key predictors of stroke, aligning with findings from the unsupervised clustering approach in high-risk groups. Additionally, previously unidentified risk factors such as elbow joint thickness, fructosamine, hemoglobin, and calcium level demonstrate potential for stroke prediction. In conclusion, machine learning facilitated accurate stroke risk predictions and highlighted potential biomarkers, offering a data-driven framework for risk assessment and biomarker discovery.
- Open Access
- 7 April 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Disease Prediction Using Machine Learning on Smartphone-Based Eye, Skin, and Voice Data: Scoping Review (Preprint)
BACKGROUND The application of machine learning methods to data generated by ubiquitous devices like smartphones presents an opportunity to enhance the quality of health care and diagnostics. Smartphones are ideal for gathering data easily, providing quick feedback on diagnoses, and proposing interventions for health improvement. OBJECTIVE We reviewed the existing literature to gather studies that have used machine learning models with smartphone-derived data for the prediction and diagnosis of health anomalies. We divided the studies into those that used machine learning models by conducting experiments to retrieve data and predict diseases, and those that used machine learning models on publicly available databases. The details of databases, experiments, and machine learning models are intended to help researchers working in the fields of machine learning and artificial intelligence in the health care domain. Researchers can use the information to design their experiments or determine the databases they could analyze. METHODS A comprehensive search of the PubMed and IEEE Xplore databases was conducted, and an in-house keyword screening method was used to filter the articles based on the content of their titles and abstracts. Subsequently, studies related to the 3 areas of voice, skin, and eye were selected and analyzed based on how data for machine learning models were extracted (ie, the use of publicly available databases or through experiments). The machine learning methods used in each study were also noted. RESULTS A total of 49 studies were identified as being relevant to the topic of interest, and among these studies, there were 31 different databases and 24 different machine learning methods. CONCLUSIONS The results provide a better understanding of how smartphone data are collected for predicting different diseases and what kinds of machine learning methods are used on these data. Similarly, publicly available databases having smartphone-based data that can be used for the diagnosis of various diseases have been presented. Our screening method could be used or improved in future studies, and our findings could be used as a reference to conduct similar studies, experiments, or statistical analyses.
- 7 April 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDisease Prediction Using Machine Learning on Smartphone-Based Eye, Skin, and Voice Data: Scoping Review (Preprint)
BACKGROUND The application of machine learning methods to data generated by ubiquitous devices like smartphones presents an opportunity to enhance the quality of health care and diagnostics. Smartphones are ideal for gathering data easily, providing quick feedback on diagnoses, and proposing interventions for health improvement. OBJECTIVE We reviewed the existing literature to gather studies that have used machine learning models with smartphone-derived data for the prediction and diagnosis of health anomalies. We divided the studies into those that used machine learning models by conducting experiments to retrieve data and predict diseases, and those that used machine learning models on publicly available databases. The details of databases, experiments, and machine learning models are intended to help researchers working in the fields of machine learning and artificial intelligence in the health care domain. Researchers can use the information to design their experiments or determine the databases they could analyze. METHODS A comprehensive search of the PubMed and IEEE Xplore databases was conducted, and an in-house keyword screening method was used to filter the articles based on the content of their titles and abstracts. Subsequently, studies related to the 3 areas of voice, skin, and eye were selected and analyzed based on how data for machine learning models were extracted (ie, the use of publicly available databases or through experiments). The machine learning methods used in each study were also noted. RESULTS A total of 49 studies were identified as being relevant to the topic of interest, and among these studies, there were 31 different databases and 24 different machine learning methods. CONCLUSIONS The results provide a better understanding of how smartphone data are collected for predicting different diseases and what kinds of machine learning methods are used on these data. Similarly, publicly available databases having smartphone-based data that can be used for the diagnosis of various diseases have been presented. Our screening method could be used or improved in future studies, and our findings could be used as a reference to conduct similar studies, experiments, or statistical analyses.
- Open Access
Predicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables
Cardiovascular disease (CVD) is one of the primary causes of death around the world. This study aimed to identify risk factors associated with CVD mortality using data from the National Health and Nutrition Examination Survey (NHANES). We created three models focusing on dietary data, non-diet-related health data, and a combination of both. Machine learning (ML) models, particularly the random forest algorithm, demonstrated robust consistency across health, nutrition, and mixed categories in predicting death from CVD. Shapley additive explanation (SHAP) values showed age, systolic blood pressure, and several other health factors as crucial variables, while fiber, calcium, and vitamin E, among others, were significant nutritional variables. Our research emphasizes the importance of comprehensive health evaluation and dietary intake in predicting CVD mortality. The inclusion of nutrition variables improved the performance of our models, underscoring the utility of dietary intake in ML-based data analysis. Further investigation using large datasets with recurring dietary recalls is necessary to enhance the effectiveness and interpretability of such models.
Open AccessPredicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables
Cardiovascular disease (CVD) is one of the primary causes of death around the world. This study aimed to identify risk factors associated with CVD mortality using data from the National Health and Nutrition Examination Survey (NHANES). We created three models focusing on dietary data, non-diet-related health data, and a combination of both. Machine learning (ML) models, particularly the random forest algorithm, demonstrated robust consistency across health, nutrition, and mixed categories in predicting death from CVD. Shapley additive explanation (SHAP) values showed age, systolic blood pressure, and several other health factors as crucial variables, while fiber, calcium, and vitamin E, among others, were significant nutritional variables. Our research emphasizes the importance of comprehensive health evaluation and dietary intake in predicting CVD mortality. The inclusion of nutrition variables improved the performance of our models, underscoring the utility of dietary intake in ML-based data analysis. Further investigation using large datasets with recurring dietary recalls is necessary to enhance the effectiveness and interpretability of such models.
- Open Access
- 13 June 2023
- Published by Wolters Kluwer Health in Pediatric Critical Care Medicine
- Vol. 24 (10) , 818-828
Combined Therapeutic Plasma Exchange and Continuous Renal Replacement Therapy in Children With Dengue-Associated Acute Liver Failure and Shock Syndrome: Single-Center Cohort From Vietnam*
OBJECTIVES: Pediatric acute liver failure (PALF) is a fatal complication in patients with severe dengue. To date, clinical data on the combination of therapeutic plasma exchange (TPE) and continuous renal replacement therapy (CRRT) for managing dengue-associated PALF concomitant with shock syndrome are limited. DESIGN: Retrospective cohort study (January 2013 to June 2022). PATIENTS: Thirty-four children. SETTING: PICU of tertiary Children’s Hospital No. 2 in Vietnam. INTERVENTIONS: We assessed a before-versus-after practice change at our center of using combined TPE and CRRT (2018 to 2022) versus CRRT alone (2013 to 2017) in managing children with dengue-associated acute liver failure and shock syndrome. Clinical and laboratory data were reviewed from PICU admission, before and 24 h after CRRT and TPE treatments. The main study outcomes were 28-day in-hospital mortality, hemodynamics, clinical hepatoencephalopathy, and liver function normalization. MEASUREMENTS AND MAIN RESULTS: A total of 34 children with a median age of 10 years (interquartile range: 7–11 yr) underwent standard-volume TPE and/or CRRT treatments. Combined TPE and CRRT (n = 19), versus CRRT alone (n = 15), was associated with lower proportion of mortality 7 of 19 (37%) versus 13 of 15 (87%), difference 50% (95% CI, 22–78; p < 0.01). Use of combined TPE and CRRT was associated with substantial advancements in clinical hepatoencephalopathy, liver transaminases, coagulation profiles, and blood lactate and ammonia levels (all p values < 0.001). CONCLUSIONS: In our experience of children with dengue-associated PALF and shock syndrome, combined use of TPE and CRRT, versus CRRT alone, is associated with better outcomes. Such combination intervention was associated with normalization of liver function, neurological status, and biochemistry. In our center we continue to use combined TPE and CRRT rather than CRRT alone.
- 13 June 2023
- Published by Wolters Kluwer Health in Pediatric Critical Care Medicine
- Vol. 24 (10) , 818-828
Open AccessCombined Therapeutic Plasma Exchange and Continuous Renal Replacement Therapy in Children With Dengue-Associated Acute Liver Failure and Shock Syndrome: Single-Center Cohort From Vietnam*
OBJECTIVES: Pediatric acute liver failure (PALF) is a fatal complication in patients with severe dengue. To date, clinical data on the combination of therapeutic plasma exchange (TPE) and continuous renal replacement therapy (CRRT) for managing dengue-associated PALF concomitant with shock syndrome are limited. DESIGN: Retrospective cohort study (January 2013 to June 2022). PATIENTS: Thirty-four children. SETTING: PICU of tertiary Children’s Hospital No. 2 in Vietnam. INTERVENTIONS: We assessed a before-versus-after practice change at our center of using combined TPE and CRRT (2018 to 2022) versus CRRT alone (2013 to 2017) in managing children with dengue-associated acute liver failure and shock syndrome. Clinical and laboratory data were reviewed from PICU admission, before and 24 h after CRRT and TPE treatments. The main study outcomes were 28-day in-hospital mortality, hemodynamics, clinical hepatoencephalopathy, and liver function normalization. MEASUREMENTS AND MAIN RESULTS: A total of 34 children with a median age of 10 years (interquartile range: 7–11 yr) underwent standard-volume TPE and/or CRRT treatments. Combined TPE and CRRT (n = 19), versus CRRT alone (n = 15), was associated with lower proportion of mortality 7 of 19 (37%) versus 13 of 15 (87%), difference 50% (95% CI, 22–78; p < 0.01). Use of combined TPE and CRRT was associated with substantial advancements in clinical hepatoencephalopathy, liver transaminases, coagulation profiles, and blood lactate and ammonia levels (all p values < 0.001). CONCLUSIONS: In our experience of children with dengue-associated PALF and shock syndrome, combined use of TPE and CRRT, versus CRRT alone, is associated with better outcomes. Such combination intervention was associated with normalization of liver function, neurological status, and biochemistry. In our center we continue to use combined TPE and CRRT rather than CRRT alone.