Proadpran Punyabukkana
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Proadpran Punyabukkana has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Atiwong Suchato
Chulalongkorn University
10 shared publications - Dittaya Wanvarie
Chulalongkorn University
4 shared publications - Ekapol Chuangsuwanich
Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand
4 shared publications - Ploy Naruemon Pratanwanich
Chulalongkorn University
3 shared publications Atiwong Suchato
2 shared publications- Chaipat Chunharas
Cognitive Clinical and Computational Neuroscience Center of Excellence, Department of Medicine, King Chulalongkorn Memorial Hospital, Chulalongkorn University
2 shared publications
Latest Publications
- Open Access
- 12 October 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Determinants of Use, Affordability, and Stakeholder Requirements for Decision Support Allocating Home-Based Laboratory Services to Older Adults With Chronic Disease in Thailand: Convergent Mixed Methods Study (Preprint)
BACKGROUND Home-based laboratory services (HBLS), in which specimens are collected at home for analysis in a hospital laboratory, are expanding in aging societies. Who uses them, what patients can pay, and what professionals require before trusting a tool that allocates them remain poorly described in middle-income Asian health systems. OBJECTIVE We aimed to (1) identify determinants of HBLS use within Andersen's Behavioral Model, (2) quantify users' preferences, travel burden, and acceptable fee, (3) elicit stakeholder requirements for a clinical decision support system (CDSS) allocating HBLS, and (4) integrate the strands into design requirements. METHODS This convergent mixed methods study at a tertiary referral center in Bangkok, Thailand, analyzed 95,818 laboratory records of 12,458 adults with chronic disease (67% aged ≥60 years; 2021-2022) and modeled ever-use of HBLS (3946 users vs 8512 never-users) by multivariable logistic regression on predisposing, enabling, and need factors. A survey of 400 users rated 20 decision factors and reported testing frequency, travel cost, and acceptable fee. Twenty semistructured interviews (8 physicians, 7 medical technologists, 3 hospital administrators, and 2 private-sector executives) underwent reflexive thematic analysis. Strands were integrated through a joint display coded for convergence, complementarity, expansion, and divergence. RESULTS Need factors had the largest adjusted odds ratios (aORs): 9.42 (95% CI 7.32-12.11) for being bedbound, 5.97 (95% CI 5.18-6.87) for being homebound, 4.67 (95% CI 3.93-5.56) for ≥3 comorbidities, and 4.18 (95% CI 3.44-5.09) for age ≥80 years. Taxi reliance (aOR 5.83, 95% CI 5.02-6.76), not working (aOR 5.21, 95% CI 4.57-5.95), and a caregiver at home (aOR 2.47, 95% CI 2.17-2.83) increased use; Social Security Scheme coverage reduced it (aOR 0.13, 95% CI 0.11-0.16) vs the Universal Coverage Scheme; estimates were similar in patients aged ≥60 years. Survey respondents were mostly older (89.0% aged >60 years) and low-income (79.0% below 10,000 Thai baht [THB] monthly); the factors most often rated most important were age (69.5%), inability to care for oneself (62.0%), and difficulty traveling (60.3%). Over half (58.0%) spent >500 THB per visit on travel, 69.7% (of 393) named 300-500 THB as an acceptable fee, and 97.0% intended to reuse the service. Six themes emerged: all groups agreed that stable chronic disease monitoring is the target, but physicians required advisory, overridable recommendations while technologists required nonnegotiable specimen rules. Integration yielded 4 meta-inferences and 12 design requirements, including 2-tier governance: a hard specimen-safety gate and an advisory patient-selection layer. CONCLUSIONS HBLS use is driven by functional dependency and multimorbidity rather than demographic convenience, concentrates among low-income older adults whose travel costs exceed the fee they find acceptable, and is trusted by professionals when specimen integrity is nonnegotiable and clinical judgment is preserved. These findings supply eligibility, financing, and governance requirements for CDSSs allocating home-based diagnostics to older adults.
- 12 October 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDeterminants of Use, Affordability, and Stakeholder Requirements for Decision Support Allocating Home-Based Laboratory Services to Older Adults With Chronic Disease in Thailand: Convergent Mixed Methods Study (Preprint)
BACKGROUND Home-based laboratory services (HBLS), in which specimens are collected at home for analysis in a hospital laboratory, are expanding in aging societies. Who uses them, what patients can pay, and what professionals require before trusting a tool that allocates them remain poorly described in middle-income Asian health systems. OBJECTIVE We aimed to (1) identify determinants of HBLS use within Andersen's Behavioral Model, (2) quantify users' preferences, travel burden, and acceptable fee, (3) elicit stakeholder requirements for a clinical decision support system (CDSS) allocating HBLS, and (4) integrate the strands into design requirements. METHODS This convergent mixed methods study at a tertiary referral center in Bangkok, Thailand, analyzed 95,818 laboratory records of 12,458 adults with chronic disease (67% aged ≥60 years; 2021-2022) and modeled ever-use of HBLS (3946 users vs 8512 never-users) by multivariable logistic regression on predisposing, enabling, and need factors. A survey of 400 users rated 20 decision factors and reported testing frequency, travel cost, and acceptable fee. Twenty semistructured interviews (8 physicians, 7 medical technologists, 3 hospital administrators, and 2 private-sector executives) underwent reflexive thematic analysis. Strands were integrated through a joint display coded for convergence, complementarity, expansion, and divergence. RESULTS Need factors had the largest adjusted odds ratios (aORs): 9.42 (95% CI 7.32-12.11) for being bedbound, 5.97 (95% CI 5.18-6.87) for being homebound, 4.67 (95% CI 3.93-5.56) for ≥3 comorbidities, and 4.18 (95% CI 3.44-5.09) for age ≥80 years. Taxi reliance (aOR 5.83, 95% CI 5.02-6.76), not working (aOR 5.21, 95% CI 4.57-5.95), and a caregiver at home (aOR 2.47, 95% CI 2.17-2.83) increased use; Social Security Scheme coverage reduced it (aOR 0.13, 95% CI 0.11-0.16) vs the Universal Coverage Scheme; estimates were similar in patients aged ≥60 years. Survey respondents were mostly older (89.0% aged >60 years) and low-income (79.0% below 10,000 Thai baht [THB] monthly); the factors most often rated most important were age (69.5%), inability to care for oneself (62.0%), and difficulty traveling (60.3%). Over half (58.0%) spent >500 THB per visit on travel, 69.7% (of 393) named 300-500 THB as an acceptable fee, and 97.0% intended to reuse the service. Six themes emerged: all groups agreed that stable chronic disease monitoring is the target, but physicians required advisory, overridable recommendations while technologists required nonnegotiable specimen rules. Integration yielded 4 meta-inferences and 12 design requirements, including 2-tier governance: a hard specimen-safety gate and an advisory patient-selection layer. CONCLUSIONS HBLS use is driven by functional dependency and multimorbidity rather than demographic convenience, concentrates among low-income older adults whose travel costs exceed the fee they find acceptable, and is trusted by professionals when specimen integrity is nonnegotiable and clinical judgment is preserved. These findings supply eligibility, financing, and governance requirements for CDSSs allocating home-based diagnostics to older adults.
- Open Access
- 5 December 2025
- Published by Springer Nature in Research Square
Enhancing Parkinson’s Diagnosis with PhonatoryVoice Analysis: Optimizing Machine LearningFeature Selection for Support Vector Machines
Parkinson’s disease (PD) is a common neurological disorder that impairs mobility and significantly impacts patients’ quality of life. Early diagnosis of PD is crucial for timely intervention and management. This study explores the potential of computer-aided analysis, specifically machine learning, for early PD diagnosis, focusing on phonatory voice analysis using dysphonia measures from sustained vowels. To enhance PD diagnosis performances, this study investigates the effectiveness of two feature selection methods, L1-norm support vector machine (L1-norm SVM) and Recursive Feature Elimination (RFE), within the framework of Support Vector Machine (SVM) and data from UCI Machine Learning Repository and Chulalongkorn Centre of Excellence for Parkinson’s Disease \& Related Disorders, a leading medical center for treating PD patients in Thailand, were utilized. Furthermore, this research uses PD voice samples to analyze dysphonia features for diagnosing PD. The results suggest that both L1-norm SVM and RFE enhance the accuracy of PD diagnosis, with L1-norm SVM showing higher performance, achieving an accuracy of 95.97%, a positive recall score of 97.16%, and an F1-macro score of 95.69%. The study also highlights the importance of key dysphonia measures, including RPDE, DFA, and HNR, in differentiating PD voices from non-PD voices. Further research is needed to explore additional feature selection methods, audio features, and datasets, which would help uncover more characteristics in early PD voice analysis and further enhance the performance of PD diagnosis.
- 5 December 2025
- Published by Springer Nature in Research Square
Open AccessEnhancing Parkinson’s Diagnosis with PhonatoryVoice Analysis: Optimizing Machine LearningFeature Selection for Support Vector Machines
Parkinson’s disease (PD) is a common neurological disorder that impairs mobility and significantly impacts patients’ quality of life. Early diagnosis of PD is crucial for timely intervention and management. This study explores the potential of computer-aided analysis, specifically machine learning, for early PD diagnosis, focusing on phonatory voice analysis using dysphonia measures from sustained vowels. To enhance PD diagnosis performances, this study investigates the effectiveness of two feature selection methods, L1-norm support vector machine (L1-norm SVM) and Recursive Feature Elimination (RFE), within the framework of Support Vector Machine (SVM) and data from UCI Machine Learning Repository and Chulalongkorn Centre of Excellence for Parkinson’s Disease \& Related Disorders, a leading medical center for treating PD patients in Thailand, were utilized. Furthermore, this research uses PD voice samples to analyze dysphonia features for diagnosing PD. The results suggest that both L1-norm SVM and RFE enhance the accuracy of PD diagnosis, with L1-norm SVM showing higher performance, achieving an accuracy of 95.97%, a positive recall score of 97.16%, and an F1-macro score of 95.69%. The study also highlights the importance of key dysphonia measures, including RPDE, DFA, and HNR, in differentiating PD voices from non-PD voices. Further research is needed to explore additional feature selection methods, audio features, and datasets, which would help uncover more characteristics in early PD voice analysis and further enhance the performance of PD diagnosis.
- Open Access
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 90-97
Pioneering Inclusion: A Low-Cost, Low-Tech Approach to Educating Thailand's First Blind Engineering Student
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 90-97
Open AccessPioneering Inclusion: A Low-Cost, Low-Tech Approach to Educating Thailand's First Blind Engineering Student
- 23 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 158-162
Realizing Global Citizen Solutions for Sustainable Development Goals: A Collaborative Engineering Course Between Thai and Japanese Students
One of the problems faced by engineering and science students is focusing solely on technical knowledge while needing more cultural context. In today's interconnected world, addressing global challenges necessitates collaborative efforts from individuals across nations. Accordingly, a joint course between Chulalongkorn University in Thailand and the Tokyo Institute of Technology in Japan has been established. The recent theme of the course, “Gamification towards Sustainable Development Goals (SDGs),” aimed to foster global citizenship and address SDGs through multicultural collaboration. This study examines the similarities and differences in problem identification between Thai and Japanese student groups participating in the course. Data was collected from discussions, presentations, and group assignments throughout the course. The findings indicate that while both groups share common concerns, such as waste management, language barriers, and transportation issues, their perspectives and proposed solutions are influenced by cultural characteristics. Thai students emphasized infrastructural inadequacies and local traditions, whereas Japanese students focused on efficiency and the impacts of tourism. The study highlights the importance of understanding cultural nuances in collaborative international projects to achieve effective and innovative solutions to global challenges.
- 23 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 158-162
Realizing Global Citizen Solutions for Sustainable Development Goals: A Collaborative Engineering Course Between Thai and Japanese Students
One of the problems faced by engineering and science students is focusing solely on technical knowledge while needing more cultural context. In today's interconnected world, addressing global challenges necessitates collaborative efforts from individuals across nations. Accordingly, a joint course between Chulalongkorn University in Thailand and the Tokyo Institute of Technology in Japan has been established. The recent theme of the course, “Gamification towards Sustainable Development Goals (SDGs),” aimed to foster global citizenship and address SDGs through multicultural collaboration. This study examines the similarities and differences in problem identification between Thai and Japanese student groups participating in the course. Data was collected from discussions, presentations, and group assignments throughout the course. The findings indicate that while both groups share common concerns, such as waste management, language barriers, and transportation issues, their perspectives and proposed solutions are influenced by cultural characteristics. Thai students emphasized infrastructural inadequacies and local traditions, whereas Japanese students focused on efficiency and the impacts of tourism. The study highlights the importance of understanding cultural nuances in collaborative international projects to achieve effective and innovative solutions to global challenges.
- Top Cited PapersOpen Access
- 25 July 2023
- Published by Elsevier in Computers and Education: Artificial Intelligence
- Vol. 5
English grammar multiple-choice question generation using Text-to-Text Transfer Transformer
- 25 July 2023
- Published by Elsevier in Computers and Education: Artificial Intelligence
- Vol. 5
Top Cited PapersOpen AccessEnglish grammar multiple-choice question generation using Text-to-Text Transfer Transformer
- 10 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 10 (18) , 15943-15952
RRWaveNet: A Compact End-to-End Multiscale Residual CNN for Robust PPG Respiratory Rate Estimation
Respiratory rate (RR) is an important biomarker as RR changes can reflect severe medical events such as heart disease, lung disease, and sleep disorders. Unfortunately, standard manual RR counting is prone to human error and cannot be performed continuously. This study proposes a method for continuously estimating RR, RRWaveNet. The method is a compact end-to-end deep learning model which does not require feature engineering and can use low-cost raw photoplethysmography (PPG) as input signal. RRWaveNet was tested subject-independently and compared to baseline in four datasets (BIDMC, CapnoBase, WESAD, and SensAI) and using three window sizes (16, 32, and 64 seconds). RRWaveNet outperformed current state-of-the-art methods with mean absolute errors at optimal window size of 1.66 1 1.01, 1.59 1 1.08, 1.92 1 0.96, and 1.23 1 0.61 breaths per minute for each dataset. In remote monitoring settings, such as in the WESAD and SensAI datasets, we apply transfer learning to improve the performance using two other ICU datasets as pretraining datasets, reducing the MAE by up to 21%. This shows that this model allows accurate and practical estimation of RR on affordable and wearable devices. Our study also shows feasibility of remote RR monitoring in the context of telemedicine and at home.
- 10 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 10 (18) , 15943-15952
RRWaveNet: A Compact End-to-End Multiscale Residual CNN for Robust PPG Respiratory Rate Estimation
Respiratory rate (RR) is an important biomarker as RR changes can reflect severe medical events such as heart disease, lung disease, and sleep disorders. Unfortunately, standard manual RR counting is prone to human error and cannot be performed continuously. This study proposes a method for continuously estimating RR, RRWaveNet. The method is a compact end-to-end deep learning model which does not require feature engineering and can use low-cost raw photoplethysmography (PPG) as input signal. RRWaveNet was tested subject-independently and compared to baseline in four datasets (BIDMC, CapnoBase, WESAD, and SensAI) and using three window sizes (16, 32, and 64 seconds). RRWaveNet outperformed current state-of-the-art methods with mean absolute errors at optimal window size of 1.66 1 1.01, 1.59 1 1.08, 1.92 1 0.96, and 1.23 1 0.61 breaths per minute for each dataset. In remote monitoring settings, such as in the WESAD and SensAI datasets, we apply transfer learning to improve the performance using two other ICU datasets as pretraining datasets, reducing the MAE by up to 21%. This shows that this model allows accurate and practical estimation of RR on affordable and wearable devices. Our study also shows feasibility of remote RR monitoring in the context of telemedicine and at home.
- Open Access
RRWaveNet: A Compact End-to-End Multi-Scale Residual CNN for Robust PPG Respiratory Rate Estimation
Respiratory rate (RR) is an important biomarker as RR changes can reflect severe medical events such as heart disease, lung disease, and sleep disorders. Unfortunately, standard manual RR counting is prone to human error and cannot be performed continuously. This study proposes a method for continuously estimating RR, RRWaveNet. The method is a compact end-to-end deep learning model which does not require feature engineering and can use low-cost raw photoplethysmography (PPG) as input signal. RRWaveNet was tested subject-independently and compared to baseline in four datasets (BIDMC, CapnoBase, WESAD, and SensAI) and using three window sizes (16, 32, and 64 seconds). RRWaveNet outperformed current state-of-the-art methods with mean absolute errors at optimal window size of 1.66 \pm 1.01, 1.59 \pm 1.08, 1.92 \pm 0.96 and 1.23 \pm 0.61 breaths per minute for each dataset. In remote monitoring settings, such as in the WESAD and SensAI datasets, we apply transfer learning to improve the performance using two other ICU datasets as pretraining datasets, reducing the MAE by up to 21$\%$. This shows that this model allows accurate and practical estimation of RR on affordable and wearable devices. Our study also shows feasibility of remote RR monitoring in the context of telemedicine and at home.
Open AccessRRWaveNet: A Compact End-to-End Multi-Scale Residual CNN for Robust PPG Respiratory Rate Estimation
Respiratory rate (RR) is an important biomarker as RR changes can reflect severe medical events such as heart disease, lung disease, and sleep disorders. Unfortunately, standard manual RR counting is prone to human error and cannot be performed continuously. This study proposes a method for continuously estimating RR, RRWaveNet. The method is a compact end-to-end deep learning model which does not require feature engineering and can use low-cost raw photoplethysmography (PPG) as input signal. RRWaveNet was tested subject-independently and compared to baseline in four datasets (BIDMC, CapnoBase, WESAD, and SensAI) and using three window sizes (16, 32, and 64 seconds). RRWaveNet outperformed current state-of-the-art methods with mean absolute errors at optimal window size of 1.66 \pm 1.01, 1.59 \pm 1.08, 1.92 \pm 0.96 and 1.23 \pm 0.61 breaths per minute for each dataset. In remote monitoring settings, such as in the WESAD and SensAI datasets, we apply transfer learning to improve the performance using two other ICU datasets as pretraining datasets, reducing the MAE by up to 21$\%$. This shows that this model allows accurate and practical estimation of RR on affordable and wearable devices. Our study also shows feasibility of remote RR monitoring in the context of telemedicine and at home.
- Open Access
- 15 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 132144-132152
Adaptive Image Preprocessing and Augmentation for Tuberculosis Screening on Out-of-Domain Chest X-Ray Dataset
Research on deep learning models for chest radiology applications have been getting much attention by the community. These works mostly focus on developing models using in-domain data. However, real-world scenarios might not always match the training set, which makes some models perform worse at the deployment stage. This work focuses on studying the effect of dataset mismatch on chest radiography and methods that can mitigate this problem. We developed the lung balance contrast enhancement technique (lung BCET) which automatically identifies the lung region and normalizes the image accordingly to improve the robustness on out-of-domain data. We also explored augmentation methods that are suitable for chest radiography. We compiled data from multiple tuberculosis (TB) datasets to evaluate and compare the performance of the preprocessing and augmentation methods using the area under the receiver operating characteristic curve (AUC) and heatmap quality. On out-of-domain testing conditions, the lung BCET preprocessing method achieves the highest AUC scores of 0.7978 and 0.6240 for the Maesot and Bureau of TB (BT) datasets, respectively. We also found that lung BCET can also be used to perform data augmentation in conjunction with standard augmentation techniques to improve the performance in both in-and out-of-domain conditions.
- 15 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 132144-132152
Open AccessAdaptive Image Preprocessing and Augmentation for Tuberculosis Screening on Out-of-Domain Chest X-Ray Dataset
Research on deep learning models for chest radiology applications have been getting much attention by the community. These works mostly focus on developing models using in-domain data. However, real-world scenarios might not always match the training set, which makes some models perform worse at the deployment stage. This work focuses on studying the effect of dataset mismatch on chest radiography and methods that can mitigate this problem. We developed the lung balance contrast enhancement technique (lung BCET) which automatically identifies the lung region and normalizes the image accordingly to improve the robustness on out-of-domain data. We also explored augmentation methods that are suitable for chest radiography. We compiled data from multiple tuberculosis (TB) datasets to evaluate and compare the performance of the preprocessing and augmentation methods using the area under the receiver operating characteristic curve (AUC) and heatmap quality. On out-of-domain testing conditions, the lung BCET preprocessing method achieves the highest AUC scores of 0.7978 and 0.6240 for the Maesot and Bureau of TB (BT) datasets, respectively. We also found that lung BCET can also be used to perform data augmentation in conjunction with standard augmentation techniques to improve the performance in both in-and out-of-domain conditions.
- Open Access
- 4 November 2022
- Published by Elsevier in The Lancet Regional Health - Southeast Asia
- Vol. 8, 100106
Breakthrough infections, hospital admissions, and mortality after major COVID-19 vaccination profiles: a prospective cohort study
- 4 November 2022
- Published by Elsevier in The Lancet Regional Health - Southeast Asia
- Vol. 8, 100106
Open AccessBreakthrough infections, hospital admissions, and mortality after major COVID-19 vaccination profiles: a prospective cohort study