Yu-Shiang Lin
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
Yu-Shiang Lin has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
Li-Yun Chen
In-Service Master Program in Artificial Intelligence in Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
4 shared publicationsAlexander Valley Chang
In-Service Master Program in Artificial Intelligence in Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
2 shared publications- Chia-Ming Lin
Professional Master Program in Artificial Intelligence in Medicine, College of Medicine, Taipei Medical University, Taipei 11031, Taiwan
2 shared publications Gina Jinna Chen
Department of Electronic and Electrical Engineering, Southern University of Science and Technology, Shenzhen, People’s Republic of China
2 shared publications- Hsiao-Cheng Chang
In-Service Master Program in Artificial Intelligence in Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
2 shared publications Yen-Tin Chen
In-Service Master Program in Artificial Intelligence in Medicine, College of Medicine, Taipei Medical University, Taipei City, Taiwan
2 shared publications
Latest Publications
- Open Access
- 12 September 2026
- Published by Springer Nature in BMC Oral Health
Machine learning-based prediction of intraoperative desaturation risk in pediatric dental sedation using preoperative data
- 12 September 2026
- Published by Springer Nature in BMC Oral Health
Open AccessMachine learning-based prediction of intraoperative desaturation risk in pediatric dental sedation using preoperative data
- Open Access
- 1 September 2026
- Published by Taylor & Francis in Journal of Multidisciplinary Healthcare
- Vol. ume 19, 1-16
Early Prediction of Postoperative Urinary Retention After General Anesthesia Using Explainable Machine Learning
Purpose: Postoperative urinary retention (POUR) is a common and clinically important perioperative complication associated with pharmacological effects and autonomic dysfunction, which may lead to increased patient discomfort and delayed recovery. Its multifactorial etiology makes early clinical identification challenging. This study investigated the feasibility of using machine learning to predict early POUR after general anesthesia using readily available, non-invasive perioperative data. Model interpretation techniques were further applied to improve transparency and identify key clinical features contributing to risk stratification. Materials and Methods: This retrospective single-center cohort study included 522 adult inpatients from Cathay General Hospital who underwent surgery under general anesthesia with endotracheal intubation and were subsequently admitted to the post-anesthesia care unit (PACU). Early POUR was defined as postoperative urinary retention occurring during the PACU stay, generally within approximately one hour after extubation. A total of 11 routinely available non-invasive clinical features were used to develop and evaluate multiple machine learning models for predicting early POUR. These machine learning models included conventional classifiers, neural network–based models, and ensemble learning algorithms, including LightGBM, a gradient boosting–based ensemble learning algorithm. SHapley Additive exPlanations (SHAP) analysis was performed to interpret model predictions and identify key clinical features associated with early POUR. Results: Early POUR occurred in 58 of 522 patients (11.1%). Among the evaluated machine learning models, LightGBM demonstrated favorable overall classification performance in predicting early POUR, with an accuracy of 81.03% and an F1 score of 81.50%. Logistic Regression achieved the highest area under the receiver operating characteristic curve (AUC) among the evaluated models (AUC = 0.7099). SHAP analysis identified several key predictors, including glycopyrrolate use, age, ASA class, anesthesia duration, and postoperative pain. Conclusion: This study suggests that machine learning may be a feasible approach for early POUR risk prediction. Interpretable approaches may facilitate individualized anesthetic planning and support clinical decision-making in the PACU, particularly in guiding early bladder monitoring and catheterization decisions for patients at risk of POUR.
- 1 September 2026
- Published by Taylor & Francis in Journal of Multidisciplinary Healthcare
- Vol. ume 19, 1-16
Open AccessEarly Prediction of Postoperative Urinary Retention After General Anesthesia Using Explainable Machine Learning
Purpose: Postoperative urinary retention (POUR) is a common and clinically important perioperative complication associated with pharmacological effects and autonomic dysfunction, which may lead to increased patient discomfort and delayed recovery. Its multifactorial etiology makes early clinical identification challenging. This study investigated the feasibility of using machine learning to predict early POUR after general anesthesia using readily available, non-invasive perioperative data. Model interpretation techniques were further applied to improve transparency and identify key clinical features contributing to risk stratification. Materials and Methods: This retrospective single-center cohort study included 522 adult inpatients from Cathay General Hospital who underwent surgery under general anesthesia with endotracheal intubation and were subsequently admitted to the post-anesthesia care unit (PACU). Early POUR was defined as postoperative urinary retention occurring during the PACU stay, generally within approximately one hour after extubation. A total of 11 routinely available non-invasive clinical features were used to develop and evaluate multiple machine learning models for predicting early POUR. These machine learning models included conventional classifiers, neural network–based models, and ensemble learning algorithms, including LightGBM, a gradient boosting–based ensemble learning algorithm. SHapley Additive exPlanations (SHAP) analysis was performed to interpret model predictions and identify key clinical features associated with early POUR. Results: Early POUR occurred in 58 of 522 patients (11.1%). Among the evaluated machine learning models, LightGBM demonstrated favorable overall classification performance in predicting early POUR, with an accuracy of 81.03% and an F1 score of 81.50%. Logistic Regression achieved the highest area under the receiver operating characteristic curve (AUC) among the evaluated models (AUC = 0.7099). SHAP analysis identified several key predictors, including glycopyrrolate use, age, ASA class, anesthesia duration, and postoperative pain. Conclusion: This study suggests that machine learning may be a feasible approach for early POUR risk prediction. Interpretable approaches may facilitate individualized anesthetic planning and support clinical decision-making in the PACU, particularly in guiding early bladder monitoring and catheterization decisions for patients at risk of POUR.
- Open Access
- 5 August 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Deep learning-based first-trimester down syndrome risk prediction in East Asian pregnant women using simplified clinical indicators
- 5 August 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessDeep learning-based first-trimester down syndrome risk prediction in East Asian pregnant women using simplified clinical indicators
- Open Access
- 1 July 2026
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 19, 1-13
Interpretable Machine Learning Models for Early Detection of Metabolic Dysfunction–Associated Steatotic Liver Disease Using Non-Invasive Routine Clinical and Laboratory Data
Introduction: The purpose of this study was to develop and evaluate multiple interpretable machine learning models for the early detection of metabolic dysfunction–associated steatotic liver disease (MASLD) using non-invasive routine clinical and laboratory data, with magnetic resonance imaging proton density fat fraction (MRI-PDFF) as a quantitative reference standard, and to assess the feasibility of MRI-PDFF–calibrated risk stratification as a pre-screening approach in clinical settings. Materials and Methods: This retrospective study analyzed a de-identified cohort of 152 patients using routine clinical and laboratory data collected at Fu Jen Catholic University Hospital (FJUH), Taiwan. A total of 17 routinely available clinical and anthropometric variables were used to develop and compare across ten interpretable machine learning models, with MRI-PDFF serving as the quantitative reference standard. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, and F1-score. Model interpretability was assessed using Shapley Additive Explanations (SHAP) at both the patient and feature levels. Results: Logistic Regression demonstrated the highest discriminative performance (AUC = 0.873). Random Forest achieved the best overall classification performance, with an accuracy of 0.816, precision of 0.803, recall of 0.814, and an F1-score of 0.807. SHAP analysis identified the Hepatic Steatosis Index (HSI), body fat composition, age, and triglycerides (TG) as the dominant contributors to MASLD risk prediction. Conclusion: This study demonstrates that an interpretable, MRI-PDFF–calibrated machine learning approach based on routinely available, non-invasive clinical data is feasible for MASLD risk stratification. Both transparent linear models and ensemble methods showed clinically meaningful performance, while SHAP analysis highlighted key metabolic and anthropometric factors contributing to risk. This approach may assist in prioritizing individuals for further imaging evaluation in clinical practice.
- 1 July 2026
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 19, 1-13
Open AccessInterpretable Machine Learning Models for Early Detection of Metabolic Dysfunction–Associated Steatotic Liver Disease Using Non-Invasive Routine Clinical and Laboratory Data
Introduction: The purpose of this study was to develop and evaluate multiple interpretable machine learning models for the early detection of metabolic dysfunction–associated steatotic liver disease (MASLD) using non-invasive routine clinical and laboratory data, with magnetic resonance imaging proton density fat fraction (MRI-PDFF) as a quantitative reference standard, and to assess the feasibility of MRI-PDFF–calibrated risk stratification as a pre-screening approach in clinical settings. Materials and Methods: This retrospective study analyzed a de-identified cohort of 152 patients using routine clinical and laboratory data collected at Fu Jen Catholic University Hospital (FJUH), Taiwan. A total of 17 routinely available clinical and anthropometric variables were used to develop and compare across ten interpretable machine learning models, with MRI-PDFF serving as the quantitative reference standard. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, and F1-score. Model interpretability was assessed using Shapley Additive Explanations (SHAP) at both the patient and feature levels. Results: Logistic Regression demonstrated the highest discriminative performance (AUC = 0.873). Random Forest achieved the best overall classification performance, with an accuracy of 0.816, precision of 0.803, recall of 0.814, and an F1-score of 0.807. SHAP analysis identified the Hepatic Steatosis Index (HSI), body fat composition, age, and triglycerides (TG) as the dominant contributors to MASLD risk prediction. Conclusion: This study demonstrates that an interpretable, MRI-PDFF–calibrated machine learning approach based on routinely available, non-invasive clinical data is feasible for MASLD risk stratification. Both transparent linear models and ensemble methods showed clinically meaningful performance, while SHAP analysis highlighted key metabolic and anthropometric factors contributing to risk. This approach may assist in prioritizing individuals for further imaging evaluation in clinical practice.
- Top Cited PapersOpen Access
- 1 February 2026
- Published by Taylor & Francis in Journal of Multidisciplinary Healthcare
- Vol. ume 19, 1-13
Explainable Machine Learning for Prediction of Early Postoperative Nausea and Vomiting After General Anesthesia
Purpose: Postoperative nausea and vomiting (PONV) remains one of the most common adverse effects associated with anesthesia care. This study aimed to explore the feasibility of applying machine learning models trained exclusively on routinely available non-invasive clinical indicators to predict early PONV risk. Explainable artificial intelligence techniques were also employed to identify the most influential predictors of early PONV. Patients and Methods: A retrospective dataset from Cathay General Hospital, including 927 patient cases and 16 non-invasive clinical indicators, was used to investigate early PONV risk prediction. This study evaluated the predictive performance of several traditional machine learning models, deep learning architectures, and ensemble learning methods to compare their classification capabilities. Results: Overall, the models demonstrated moderate discriminative performance. The random forest model achieved an accuracy of 83.5% with balanced precision (80.81%) and recall (83.5%), while the logistic regression model attained an AUC of 0.6905. Analysis of positive SHAP values identified the top 7 most influential predictors of early PONV. These included pharmacologic interventions (eg, neostigmine), pre-existing comorbidities (eg, history of nausea and vomiting, history of cardiovascular disease), demographic characteristics (eg, gender), postoperative pain, and anesthetic and surgical factors (eg, type of surgery and duration of anesthesia). Moreover, SHAP analysis revealed that the use of dexamethasone was negatively associated with the predicted risk in the model, suggesting its potential protective role in the prevention of early PONV. Conclusion: By generating explainable outputs, this study bridges the gap between algorithmic prediction and clinical decision-making, allowing anesthesiologists to better recognize underlying risk factors and make informed, evidence-based decisions in perioperative management.
- 1 February 2026
- Published by Taylor & Francis in Journal of Multidisciplinary Healthcare
- Vol. ume 19, 1-13
Top Cited PapersOpen AccessExplainable Machine Learning for Prediction of Early Postoperative Nausea and Vomiting After General Anesthesia
Purpose: Postoperative nausea and vomiting (PONV) remains one of the most common adverse effects associated with anesthesia care. This study aimed to explore the feasibility of applying machine learning models trained exclusively on routinely available non-invasive clinical indicators to predict early PONV risk. Explainable artificial intelligence techniques were also employed to identify the most influential predictors of early PONV. Patients and Methods: A retrospective dataset from Cathay General Hospital, including 927 patient cases and 16 non-invasive clinical indicators, was used to investigate early PONV risk prediction. This study evaluated the predictive performance of several traditional machine learning models, deep learning architectures, and ensemble learning methods to compare their classification capabilities. Results: Overall, the models demonstrated moderate discriminative performance. The random forest model achieved an accuracy of 83.5% with balanced precision (80.81%) and recall (83.5%), while the logistic regression model attained an AUC of 0.6905. Analysis of positive SHAP values identified the top 7 most influential predictors of early PONV. These included pharmacologic interventions (eg, neostigmine), pre-existing comorbidities (eg, history of nausea and vomiting, history of cardiovascular disease), demographic characteristics (eg, gender), postoperative pain, and anesthetic and surgical factors (eg, type of surgery and duration of anesthesia). Moreover, SHAP analysis revealed that the use of dexamethasone was negatively associated with the predicted risk in the model, suggesting its potential protective role in the prevention of early PONV. Conclusion: By generating explainable outputs, this study bridges the gap between algorithmic prediction and clinical decision-making, allowing anesthesiologists to better recognize underlying risk factors and make informed, evidence-based decisions in perioperative management.
- Open Access
- 1 March 2025
- Published by Taylor & Francis in Risk Management and Healthcare Policy
- Vol. ume 18, 1109-1120
Machine Learning-Based Prediction of First Trimester Down Syndrome Risk in East Asian Populations
Purpose: Down syndrome is the most common chromosomal abnormality in newborns, often leading to developmental delays and congenital structural anomalies. This study employed multiple machine learning models to perform risk prediction and result exploration for first-trimester Down syndrome in East Asian populations, aiming to identify an optimal risk prediction model that will enhance future predictions of Down syndrome risk and improve the efficiency of the screening process. Patients and Methods: This study collected data from the Down syndrome screening database at Taipei Chang Gung Memorial Hospital from May 1, 2018, to February 29, 2024. The dataset included 3,812 cases available for analysis, comprising 165 high-risk cases and 3,647 low-risk cases. Fourteen features (including maternal age, nuchal translucency thickness, serum markers, etc.) were input into the twelve machine learning models, along with seven data-balancing algorithms, to explore the risk prediction outcomes. The performance of these models was thoroughly evaluated using AUC (Area Under the Curve), accuracy, precision, recall, and F1 scores. Results: Among the twelve machine learning models, the highest recall of 0.84 for high-risk cases was achieved by LightGBM combined with the RUS (Random Undersampling) data balancing algorithm. The highest AUC of 0.939 was attained by the ANN and LSTM models when combined with the ROS (Random Oversampling) data balancing algorithm. Conclusion: The proposed ANN machine learning model, based on deep neural networks and combined with the ROS data balancing method, achieved an impressive AUC of 0.939 for classifying first-trimester Down syndrome risk in the East Asian population. Notably, this model also achieved an outstanding classification accuracy of 0.97. These results demonstrate the potential of the proposed ANN machine learning model for the accurate prediction of first-trimester Down syndrome risk.
- 1 March 2025
- Published by Taylor & Francis in Risk Management and Healthcare Policy
- Vol. ume 18, 1109-1120
Open AccessMachine Learning-Based Prediction of First Trimester Down Syndrome Risk in East Asian Populations
Purpose: Down syndrome is the most common chromosomal abnormality in newborns, often leading to developmental delays and congenital structural anomalies. This study employed multiple machine learning models to perform risk prediction and result exploration for first-trimester Down syndrome in East Asian populations, aiming to identify an optimal risk prediction model that will enhance future predictions of Down syndrome risk and improve the efficiency of the screening process. Patients and Methods: This study collected data from the Down syndrome screening database at Taipei Chang Gung Memorial Hospital from May 1, 2018, to February 29, 2024. The dataset included 3,812 cases available for analysis, comprising 165 high-risk cases and 3,647 low-risk cases. Fourteen features (including maternal age, nuchal translucency thickness, serum markers, etc.) were input into the twelve machine learning models, along with seven data-balancing algorithms, to explore the risk prediction outcomes. The performance of these models was thoroughly evaluated using AUC (Area Under the Curve), accuracy, precision, recall, and F1 scores. Results: Among the twelve machine learning models, the highest recall of 0.84 for high-risk cases was achieved by LightGBM combined with the RUS (Random Undersampling) data balancing algorithm. The highest AUC of 0.939 was attained by the ANN and LSTM models when combined with the ROS (Random Oversampling) data balancing algorithm. Conclusion: The proposed ANN machine learning model, based on deep neural networks and combined with the ROS data balancing method, achieved an impressive AUC of 0.939 for classifying first-trimester Down syndrome risk in the East Asian population. Notably, this model also achieved an outstanding classification accuracy of 0.97. These results demonstrate the potential of the proposed ANN machine learning model for the accurate prediction of first-trimester Down syndrome risk.
- Open Access
- 1 November 2024
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 17, 5163-5174
Predicting Early Treatment Effectiveness in Bell’s Palsy Using Machine Learning: A Focus on Corticosteroids and Antivirals
Purpose: Facial nerve paralysis, particularly Bell’s palsy, manifests as a rapid onset of unilateral facial weakness or paralysis. Despite most patients recovering within three to six months, a significant proportion experience poor recovery. This study utilized six machine learning models to investigate the effectiveness of early treatment in Bell’s palsy. Patients and Methods: We applied data from 17 hospitals in Scotland to predict treatment outcomes. Patients were randomized into four groups: Prednisolone (corticosteroids), Acyclovir (antivirals), both, and placebo. Outcomes, defined as full resolution of symptoms, were assessed using the House-Brackmann scale at 3 and 9 months post-treatment. We employed six different machine learning models to predict recovery outcomes and evaluated model performance using AUC, precision, recall, and F1-score. Results: Among 493 patients, 72.6% recovered after three months and 89.5% after nine months. Logistic regression demonstrated the highest predictive performance for both 3-month (AUC = 0.751) and 9-month recovery (AUC = 0.720). Additionally, several models achieved Precision levels exceeding 0.9. We further employed the best-performing logistic regression for feature ranking, indicating that the patient’s age and prednisolone administration are the most significant predictors of recovery. Conclusion: The results highlight the potential of machine learning models in predicting the effectiveness of early treatment. This study conducted a comprehensive comparison of six different machine learning models, with the logistic regression showing the highest predictive performance for both 3-month and 9-month recovery. Additionally, feature ranking using logistic regression supported the importance of Prednisolone in treatment. Notably, our findings revealed the significance of age in prognosis evaluation for the first time. This suggests that future research should further develop age-specific prognostic models, enabling clinicians to tailor individualized treatment strategies more effectively. This previously unrecognized discovery provides a foundation for prognostic analysis in Bell’s palsy patients.
- 1 November 2024
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 17, 5163-5174
Open AccessPredicting Early Treatment Effectiveness in Bell’s Palsy Using Machine Learning: A Focus on Corticosteroids and Antivirals
Purpose: Facial nerve paralysis, particularly Bell’s palsy, manifests as a rapid onset of unilateral facial weakness or paralysis. Despite most patients recovering within three to six months, a significant proportion experience poor recovery. This study utilized six machine learning models to investigate the effectiveness of early treatment in Bell’s palsy. Patients and Methods: We applied data from 17 hospitals in Scotland to predict treatment outcomes. Patients were randomized into four groups: Prednisolone (corticosteroids), Acyclovir (antivirals), both, and placebo. Outcomes, defined as full resolution of symptoms, were assessed using the House-Brackmann scale at 3 and 9 months post-treatment. We employed six different machine learning models to predict recovery outcomes and evaluated model performance using AUC, precision, recall, and F1-score. Results: Among 493 patients, 72.6% recovered after three months and 89.5% after nine months. Logistic regression demonstrated the highest predictive performance for both 3-month (AUC = 0.751) and 9-month recovery (AUC = 0.720). Additionally, several models achieved Precision levels exceeding 0.9. We further employed the best-performing logistic regression for feature ranking, indicating that the patient’s age and prednisolone administration are the most significant predictors of recovery. Conclusion: The results highlight the potential of machine learning models in predicting the effectiveness of early treatment. This study conducted a comprehensive comparison of six different machine learning models, with the logistic regression showing the highest predictive performance for both 3-month and 9-month recovery. Additionally, feature ranking using logistic regression supported the importance of Prednisolone in treatment. Notably, our findings revealed the significance of age in prognosis evaluation for the first time. This suggests that future research should further develop age-specific prognostic models, enabling clinicians to tailor individualized treatment strategies more effectively. This previously unrecognized discovery provides a foundation for prognostic analysis in Bell’s palsy patients.
- Open Access
- 25 April 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (9) , 1303
TPTM-HANN-GA: A Novel Hyperparameter Optimization Framework Integrating the Taguchi Method, an Artificial Neural Network, and a Genetic Algorithm for the Precise Prediction of Cardiovascular Disease Risk
The timely and precise prediction of cardiovascular disease (CVD) risk is essential for effective prevention and intervention. This study proposes a novel framework that integrates the two-phase Taguchi method (TPTM), the hyperparameter artificial neural network (HANN), and a genetic algorithm (GA) called TPTM-HANN-GA. This framework efficiently optimizes hyperparameters for an artificial neural network (ANN) model during the training stage, significantly enhancing prediction accuracy for cardiovascular disease (CVD) risk. The proposed TPTM-HANN-GA framework requires far fewer experiments than a traditional grid search, making it highly suitable for application in resource-constrained, low-power computers, and edge artificial intelligence (edge AI) devices. Furthermore, the proposed TPTM-HANN-GA framework successfully identified the optimal configurations for the ANN model’s hyperparameters, resulting in a hidden layer of 4 nodes, a tanh activation function, an SGD optimizer, a learning rate of 0.23425849, a momentum rate of 0.75462782, and seven hidden nodes. This optimized ANN model achieves 74.25% accuracy in predicting the risk of cardiovascular disease, which exceeds the existing state-of-the-art GA-ANN and TSTO-ANN models. The proposed TPTM-HANN-GA framework enables personalized CVD prediction to be efficiently conducted on low-power computers and edge-AI devices, achieving the goal of point-of-care testing (POCT) and empowering individuals to manage their heart health effectively.
- 25 April 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (9) , 1303
Open AccessTPTM-HANN-GA: A Novel Hyperparameter Optimization Framework Integrating the Taguchi Method, an Artificial Neural Network, and a Genetic Algorithm for the Precise Prediction of Cardiovascular Disease Risk
The timely and precise prediction of cardiovascular disease (CVD) risk is essential for effective prevention and intervention. This study proposes a novel framework that integrates the two-phase Taguchi method (TPTM), the hyperparameter artificial neural network (HANN), and a genetic algorithm (GA) called TPTM-HANN-GA. This framework efficiently optimizes hyperparameters for an artificial neural network (ANN) model during the training stage, significantly enhancing prediction accuracy for cardiovascular disease (CVD) risk. The proposed TPTM-HANN-GA framework requires far fewer experiments than a traditional grid search, making it highly suitable for application in resource-constrained, low-power computers, and edge artificial intelligence (edge AI) devices. Furthermore, the proposed TPTM-HANN-GA framework successfully identified the optimal configurations for the ANN model’s hyperparameters, resulting in a hidden layer of 4 nodes, a tanh activation function, an SGD optimizer, a learning rate of 0.23425849, a momentum rate of 0.75462782, and seven hidden nodes. This optimized ANN model achieves 74.25% accuracy in predicting the risk of cardiovascular disease, which exceeds the existing state-of-the-art GA-ANN and TSTO-ANN models. The proposed TPTM-HANN-GA framework enables personalized CVD prediction to be efficiently conducted on low-power computers and edge-AI devices, achieving the goal of point-of-care testing (POCT) and empowering individuals to manage their heart health effectively.
- Open Access
- 4 November 2023
- Published by MDPI AG in Bioengineering
- Vol. 10 (11) , 1286
Utilizing a Two-Stage Taguchi Method and Artificial Neural Network for the Precise Forecasting of Cardiovascular Disease Risk
The complexity of cardiovascular disease onset emphasizes the vital role of early detection in prevention. This study aims to enhance disease prediction accuracy using personal devices, aligning with point-of-care testing (POCT) objectives. This study introduces a two-stage Taguchi optimization (TSTO) method to boost predictive accuracy in an artificial neural network (ANN) model while minimizing computational costs. In the first stage, optimal hyperparameter levels and trends were identified. The second stage determined the best settings for the ANN model’s hyperparameters. In this study, we applied the proposed TSTO method with a personal computer to the Kaggle Cardiovascular Disease dataset. Subsequently, we identified the best setting for the hyperparameters of the ANN model, setting the hidden layer to 4, activation function to tanh, optimizer to SGD, learning rate to 0.25, momentum rate to 0.85, and hidden nodes to 10. This setting led to a state-of-the-art accuracy of 74.14% in predicting the risk of cardiovascular disease. Moreover, the proposed TSTO method significantly reduced the number of experiments by a factor of 40.5 compared to the traditional grid search method. The TSTO method accurately predicts cardiovascular risk and conserves computational resources. It is adaptable for low-power devices, aiding the goal of POCT.
- 4 November 2023
- Published by MDPI AG in Bioengineering
- Vol. 10 (11) , 1286
Open AccessUtilizing a Two-Stage Taguchi Method and Artificial Neural Network for the Precise Forecasting of Cardiovascular Disease Risk
The complexity of cardiovascular disease onset emphasizes the vital role of early detection in prevention. This study aims to enhance disease prediction accuracy using personal devices, aligning with point-of-care testing (POCT) objectives. This study introduces a two-stage Taguchi optimization (TSTO) method to boost predictive accuracy in an artificial neural network (ANN) model while minimizing computational costs. In the first stage, optimal hyperparameter levels and trends were identified. The second stage determined the best settings for the ANN model’s hyperparameters. In this study, we applied the proposed TSTO method with a personal computer to the Kaggle Cardiovascular Disease dataset. Subsequently, we identified the best setting for the hyperparameters of the ANN model, setting the hidden layer to 4, activation function to tanh, optimizer to SGD, learning rate to 0.25, momentum rate to 0.85, and hidden nodes to 10. This setting led to a state-of-the-art accuracy of 74.14% in predicting the risk of cardiovascular disease. Moreover, the proposed TSTO method significantly reduced the number of experiments by a factor of 40.5 compared to the traditional grid search method. The TSTO method accurately predicts cardiovascular risk and conserves computational resources. It is adaptable for low-power devices, aiding the goal of POCT.
- Open Access
A Novel Two-Stage Induced Deep Learning System for Classifying Similar Drugs with Diverse Packaging
Dispensing errors play a crucial role in various medical errors, unfortunately emerging as the third leading cause of death in the United States. This alarming statistic has spurred the World Health Organization (WHO) into action, leading to the initiation of the Medication Without Harm Campaign. The primary objective of this campaign is to prevent dispensing errors from occurring and ensure patient safety. Due to the rapid development of deep learning technology, there has been a significant increase in the development of automatic dispensing systems based on deep learning classification to avoid dispensing errors. However, most previous studies have focused on developing deep learning classification systems for unpackaged pills or drugs with the same type of packaging. However, in the actual dispensing process, thousands of similar drugs with diverse packaging within a healthcare facility greatly increase the risk of dispensing errors. In this study, we proposed a novel two-stage induced deep learning (TSIDL)-based system to classify similar drugs with diverse packaging efficiently. The results demonstrate that the proposed TSIDL method outperforms state-of-the-art CNN models in all classification metrics. It achieved a state-of-the-art classification accuracy of 99.39%. Moreover, this study also demonstrated that the TSIDL method achieved an inference time of only 3.12 ms per image. These results highlight the potential of real-time classification for similar drugs with diverse packaging and their applications in future dispensing systems, which can prevent dispensing errors from occurring and ensure patient safety efficiently.
Open AccessA Novel Two-Stage Induced Deep Learning System for Classifying Similar Drugs with Diverse Packaging
Dispensing errors play a crucial role in various medical errors, unfortunately emerging as the third leading cause of death in the United States. This alarming statistic has spurred the World Health Organization (WHO) into action, leading to the initiation of the Medication Without Harm Campaign. The primary objective of this campaign is to prevent dispensing errors from occurring and ensure patient safety. Due to the rapid development of deep learning technology, there has been a significant increase in the development of automatic dispensing systems based on deep learning classification to avoid dispensing errors. However, most previous studies have focused on developing deep learning classification systems for unpackaged pills or drugs with the same type of packaging. However, in the actual dispensing process, thousands of similar drugs with diverse packaging within a healthcare facility greatly increase the risk of dispensing errors. In this study, we proposed a novel two-stage induced deep learning (TSIDL)-based system to classify similar drugs with diverse packaging efficiently. The results demonstrate that the proposed TSIDL method outperforms state-of-the-art CNN models in all classification metrics. It achieved a state-of-the-art classification accuracy of 99.39%. Moreover, this study also demonstrated that the TSIDL method achieved an inference time of only 3.12 ms per image. These results highlight the potential of real-time classification for similar drugs with diverse packaging and their applications in future dispensing systems, which can prevent dispensing errors from occurring and ensure patient safety efficiently.