Myat Su Yin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Myat Su Yin has a total of 84 co-authors. Here are the top co-authors by number of shared publications:
- Peter Haddawy
University of Bremen, Bremen, Germany
10 shared publications - Siriwan Suebnukarn
Faculty of Dentistry, Thammasat University
4 shared publications - Patchara Sriwichai
Department of Medical Entomology, Faculty of Tropical Medicine, Mahidol University, Ratchawithi Road, Ratchathewi, Bangkok 10400, Thailand
3 shared publications - Andre Mühlenbrock
Faculty of Mathematics and Computer Science, University of Bremen, Bremen, Germany
2 shared publications Chaitawat Sa-Ngamuang
Faculty of ICT, Mahidol University, Nakhon Pathom, Thailand
2 shared publications- Dominique Joseph Bicout
Univ. Grenoble Alpes, CNRS, UMR 5525, VetAgro Sup, Grenoble INP, TIMC, EPSP, Grenoble, France
2 shared publications
Latest Publications
- Open Access
- 1 October 2026
- Published by JMIR Publications Inc. in JMIR Serious Games
- Vol. 14, e100884
Virtual Reality–Based Mirror Guidance for Indirect Vision Training in Minimally Invasive Endodontic Access Cavity Preparation Among Dental Students: Randomized Controlled Crossover Trial
Background: Minimally invasive endodontics requires precise visualization to preserve tooth structure, often relying on indirect vision with a dental mirror—a challenging skill for dental students. Although virtual reality (VR) simulators provide immersive environments for dental training, most existing systems primarily focus on procedural simulation and haptic feedback, with limited capabilities for real-time mirror positioning and indirect visual guidance. Objective: The aim of this study was to evaluate verbal, visual, and haptic VR-based mirror guidance for indirect vision during minimally invasive endodontic access cavity preparation. Methods: Eligible fourth-year preclinical dental students were recruited through institutional email invitations and public announcements. A randomized controlled crossover trial was conducted with 30 participants, who completed all VR sessions in a simulation laboratory under direct investigator supervision to ensure standardized implementation of the training protocol. Participants performed simulated maxillary first molar access cavity preparations under 4 VR mirror guidance conditions—verbal guidance, visual guidance, haptic guidance, and no guidance—following a randomized 4 × 4 Latin square sequence with a 48-hour washout period between sessions. The primary outcome was outcome error score. Exploratory secondary outcomes included operating area visibility, operating area obstruction, operator posture error, task completion time, and Virtual Reality System Usability Questionnaire score. Outcomes were analyzed using linear mixed-effects models accounting for the crossover design. Estimated marginal mean differences (MDs) with 95% CIs and partial eta-squared (ηp2) effect sizes were reported, with Bonferroni-adjusted pairwise comparisons used to account for multiple comparisons. Results: Guidance modality significantly affected the primary outcome, outcome error score (F3,84=20.34; P<.001; ηp2=0.421). Compared with no guidance, outcome error scores were significantly lower with verbal (MD −0.50, 95% CI −0.76 to −0.24), visual (MD −0.68, 95% CI −0.94 to −0.42), and haptic guidance (MD −0.63, 95% CI −0.89 to −0.37; all adjusted P<.001). Guidance modality also significantly affected operating area visibility (F3,84=12.31; P<.001; ηp2=0.305). Compared with no guidance, visibility duration was significantly greater with verbal (MD 37.79, 95% CI 11.15-64.43 seconds; adjusted P=.001), visual (MD 59.15, 95% CI 32.51-85.79 seconds; adjusted P<.001), and haptic guidance (MD 31.52, 95% CI 4.90-58.15 seconds; adjusted P=.01). No significant differences were detected for operating area obstruction, operator posture error, task completion time, or usability. Conclusions: VR-based mirror guidance improved the outcome error score and increased operating area visibility during minimally invasive endodontic access cavity preparation. Verbal, visual, and haptic guidance resulted in fewer outcome errors and greater operating area visibility than no guidance, while no significant differences were detected in other performance measures or usability. These findings demonstrate the potential of diverse feedback modalities to provide effective guidance for indirect vision and inform future development of adaptive VR training systems using 1 or more modalities to suit learner needs. Trial Registration: Thai Clinical Trials Registry TCTR20260126008; https://www.thaiclinicaltrials.org/show/TCTR20260126008
- 1 October 2026
- Published by JMIR Publications Inc. in JMIR Serious Games
- Vol. 14, e100884
Open AccessVirtual Reality–Based Mirror Guidance for Indirect Vision Training in Minimally Invasive Endodontic Access Cavity Preparation Among Dental Students: Randomized Controlled Crossover Trial
Background: Minimally invasive endodontics requires precise visualization to preserve tooth structure, often relying on indirect vision with a dental mirror—a challenging skill for dental students. Although virtual reality (VR) simulators provide immersive environments for dental training, most existing systems primarily focus on procedural simulation and haptic feedback, with limited capabilities for real-time mirror positioning and indirect visual guidance. Objective: The aim of this study was to evaluate verbal, visual, and haptic VR-based mirror guidance for indirect vision during minimally invasive endodontic access cavity preparation. Methods: Eligible fourth-year preclinical dental students were recruited through institutional email invitations and public announcements. A randomized controlled crossover trial was conducted with 30 participants, who completed all VR sessions in a simulation laboratory under direct investigator supervision to ensure standardized implementation of the training protocol. Participants performed simulated maxillary first molar access cavity preparations under 4 VR mirror guidance conditions—verbal guidance, visual guidance, haptic guidance, and no guidance—following a randomized 4 × 4 Latin square sequence with a 48-hour washout period between sessions. The primary outcome was outcome error score. Exploratory secondary outcomes included operating area visibility, operating area obstruction, operator posture error, task completion time, and Virtual Reality System Usability Questionnaire score. Outcomes were analyzed using linear mixed-effects models accounting for the crossover design. Estimated marginal mean differences (MDs) with 95% CIs and partial eta-squared (ηp2) effect sizes were reported, with Bonferroni-adjusted pairwise comparisons used to account for multiple comparisons. Results: Guidance modality significantly affected the primary outcome, outcome error score (F3,84=20.34; P<.001; ηp2=0.421). Compared with no guidance, outcome error scores were significantly lower with verbal (MD −0.50, 95% CI −0.76 to −0.24), visual (MD −0.68, 95% CI −0.94 to −0.42), and haptic guidance (MD −0.63, 95% CI −0.89 to −0.37; all adjusted P<.001). Guidance modality also significantly affected operating area visibility (F3,84=12.31; P<.001; ηp2=0.305). Compared with no guidance, visibility duration was significantly greater with verbal (MD 37.79, 95% CI 11.15-64.43 seconds; adjusted P=.001), visual (MD 59.15, 95% CI 32.51-85.79 seconds; adjusted P<.001), and haptic guidance (MD 31.52, 95% CI 4.90-58.15 seconds; adjusted P=.01). No significant differences were detected for operating area obstruction, operator posture error, task completion time, or usability. Conclusions: VR-based mirror guidance improved the outcome error score and increased operating area visibility during minimally invasive endodontic access cavity preparation. Verbal, visual, and haptic guidance resulted in fewer outcome errors and greater operating area visibility than no guidance, while no significant differences were detected in other performance measures or usability. These findings demonstrate the potential of diverse feedback modalities to provide effective guidance for indirect vision and inform future development of adaptive VR training systems using 1 or more modalities to suit learner needs. Trial Registration: Thai Clinical Trials Registry TCTR20260126008; https://www.thaiclinicaltrials.org/show/TCTR20260126008
- Open Access
- 11 May 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Virtual Reality–Based Mirror Guidance for Indirect Vision Training in Minimally Invasive Endodontic Access Cavity Preparation Among Dental Students: Randomized Controlled Crossover Trial (Preprint)
BACKGROUND Minimally invasive endodontics requires precise visualization to preserve tooth structure, often relying on indirect vision with a dental mirror—a challenging skill for dental students. Although virtual reality (VR) simulators provide immersive environments for dental training, most existing systems primarily focus on procedural simulation and haptic feedback, with limited capabilities for real-time mirror positioning and indirect visual guidance. OBJECTIVE The aim of this study was to evaluate verbal, visual, and haptic VR-based mirror guidance for indirect vision during minimally invasive endodontic access cavity preparation. METHODS Eligible fourth-year preclinical dental students were recruited through institutional email invitations and public announcements. A randomized controlled crossover trial was conducted with 30 participants, who completed all VR sessions in a simulation laboratory under direct investigator supervision to ensure standardized implementation of the training protocol. Participants performed simulated maxillary first molar access cavity preparations under 4 VR mirror guidance conditions—verbal guidance, visual guidance, haptic guidance, and no guidance—following a randomized 4 × 4 Latin square sequence with a 48-hour washout period between sessions. The primary outcome was outcome error score. Exploratory secondary outcomes included operating area visibility, operating area obstruction, operator posture error, task completion time, and Virtual Reality System Usability Questionnaire score. Outcomes were analyzed using linear mixed-effects models accounting for the crossover design. Estimated marginal mean differences (MDs) with 95% CIs and partial eta-squared (ηp2) effect sizes were reported, with Bonferroni-adjusted pairwise comparisons used to account for multiple comparisons. RESULTS Guidance modality significantly affected the primary outcome, outcome error score (F3,84=20.34; P<.001; ηp2=0.421). Compared with no guidance, outcome error scores were significantly lower with verbal (MD −0.50, 95% CI −0.76 to −0.24), visual (MD −0.68, 95% CI −0.94 to −0.42), and haptic guidance (MD −0.63, 95% CI −0.89 to −0.37; all adjusted P<.001). Guidance modality also significantly affected operating area visibility (F3,84=12.31; P<.001; ηp2=0.305). Compared with no guidance, visibility duration was significantly greater with verbal (MD 37.79, 95% CI 11.15-64.43 seconds; adjusted P=.001), visual (MD 59.15, 95% CI 32.51-85.79 seconds; adjusted P<.001), and haptic guidance (MD 31.52, 95% CI 4.90-58.15 seconds; adjusted P=.01). No significant differences were detected for operating area obstruction, operator posture error, task completion time, or usability. CONCLUSIONS VR-based mirror guidance improved the outcome error score and increased operating area visibility during minimally invasive endodontic access cavity preparation. Verbal, visual, and haptic guidance resulted in fewer outcome errors and greater operating area visibility than no guidance, while no significant differences were detected in other performance measures or usability. These findings demonstrate the potential of diverse feedback modalities to provide effective guidance for indirect vision and inform future development of adaptive VR training systems using 1 or more modalities to suit learner needs. CLINICALTRIAL Thai Clinical Trials Registry TCTR20260126008; https://www.thaiclinicaltrials.org/show/TCTR20260126008
- 11 May 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessVirtual Reality–Based Mirror Guidance for Indirect Vision Training in Minimally Invasive Endodontic Access Cavity Preparation Among Dental Students: Randomized Controlled Crossover Trial (Preprint)
BACKGROUND Minimally invasive endodontics requires precise visualization to preserve tooth structure, often relying on indirect vision with a dental mirror—a challenging skill for dental students. Although virtual reality (VR) simulators provide immersive environments for dental training, most existing systems primarily focus on procedural simulation and haptic feedback, with limited capabilities for real-time mirror positioning and indirect visual guidance. OBJECTIVE The aim of this study was to evaluate verbal, visual, and haptic VR-based mirror guidance for indirect vision during minimally invasive endodontic access cavity preparation. METHODS Eligible fourth-year preclinical dental students were recruited through institutional email invitations and public announcements. A randomized controlled crossover trial was conducted with 30 participants, who completed all VR sessions in a simulation laboratory under direct investigator supervision to ensure standardized implementation of the training protocol. Participants performed simulated maxillary first molar access cavity preparations under 4 VR mirror guidance conditions—verbal guidance, visual guidance, haptic guidance, and no guidance—following a randomized 4 × 4 Latin square sequence with a 48-hour washout period between sessions. The primary outcome was outcome error score. Exploratory secondary outcomes included operating area visibility, operating area obstruction, operator posture error, task completion time, and Virtual Reality System Usability Questionnaire score. Outcomes were analyzed using linear mixed-effects models accounting for the crossover design. Estimated marginal mean differences (MDs) with 95% CIs and partial eta-squared (ηp2) effect sizes were reported, with Bonferroni-adjusted pairwise comparisons used to account for multiple comparisons. RESULTS Guidance modality significantly affected the primary outcome, outcome error score (F3,84=20.34; P<.001; ηp2=0.421). Compared with no guidance, outcome error scores were significantly lower with verbal (MD −0.50, 95% CI −0.76 to −0.24), visual (MD −0.68, 95% CI −0.94 to −0.42), and haptic guidance (MD −0.63, 95% CI −0.89 to −0.37; all adjusted P<.001). Guidance modality also significantly affected operating area visibility (F3,84=12.31; P<.001; ηp2=0.305). Compared with no guidance, visibility duration was significantly greater with verbal (MD 37.79, 95% CI 11.15-64.43 seconds; adjusted P=.001), visual (MD 59.15, 95% CI 32.51-85.79 seconds; adjusted P<.001), and haptic guidance (MD 31.52, 95% CI 4.90-58.15 seconds; adjusted P=.01). No significant differences were detected for operating area obstruction, operator posture error, task completion time, or usability. CONCLUSIONS VR-based mirror guidance improved the outcome error score and increased operating area visibility during minimally invasive endodontic access cavity preparation. Verbal, visual, and haptic guidance resulted in fewer outcome errors and greater operating area visibility than no guidance, while no significant differences were detected in other performance measures or usability. These findings demonstrate the potential of diverse feedback modalities to provide effective guidance for indirect vision and inform future development of adaptive VR training systems using 1 or more modalities to suit learner needs. CLINICALTRIAL Thai Clinical Trials Registry TCTR20260126008; https://www.thaiclinicaltrials.org/show/TCTR20260126008
- Open Access
Mapping environmental suitability for Pythium insidiosum under current and future climate conditions
- Open Access
- 4 August 2025
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20 (8) , e0327360
Prognostic prediction of dengue hemorrhagic fever in pediatric patients with suspected dengue infection: A multi-site study
Dengue virus (DENV) infection is a major global health problem. While DENV infection rarely results in serious complications, the more severe illness dengue hemorrhagic fever (DHF) has a significant mortality rate due to the associated plasma leakage that may lead to hypovolemic shock. Proper care thus requires identifying patients with DHF among those with suspected dengue so that they can be provided with adequate and prompt fluid replacement. In this study we used seventeen years of pediatric patient data from a prospective cohort study in two hospitals in Thailand to develop models to predict DHF among patients with suspected dengue infection. We produced models for a general hospital setting and for a primary care unit setting lacking lab facilities. The best model using combined data from both hospitals achieved an AUC of 0.90 for the general hospital setting and 0.79 for the primary care unit setting. We then investigated the generalizability of the models by training models with data from one hospital and testing them with data from the other. For some models, we found a significant reduction in performance. Possible sources of this are differences in how attributes are defined or measured and differences in the hematological parameters of the two patient populations. We conclude that while high accuracy prediction of DHF is possible, care must be taken when applying DHF predictive models from one clinical setting to another.
- 4 August 2025
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20 (8) , e0327360
Open AccessPrognostic prediction of dengue hemorrhagic fever in pediatric patients with suspected dengue infection: A multi-site study
Dengue virus (DENV) infection is a major global health problem. While DENV infection rarely results in serious complications, the more severe illness dengue hemorrhagic fever (DHF) has a significant mortality rate due to the associated plasma leakage that may lead to hypovolemic shock. Proper care thus requires identifying patients with DHF among those with suspected dengue so that they can be provided with adequate and prompt fluid replacement. In this study we used seventeen years of pediatric patient data from a prospective cohort study in two hospitals in Thailand to develop models to predict DHF among patients with suspected dengue infection. We produced models for a general hospital setting and for a primary care unit setting lacking lab facilities. The best model using combined data from both hospitals achieved an AUC of 0.90 for the general hospital setting and 0.79 for the primary care unit setting. We then investigated the generalizability of the models by training models with data from one hospital and testing them with data from the other. For some models, we found a significant reduction in performance. Possible sources of this are differences in how attributes are defined or measured and differences in the hematological parameters of the two patient populations. We conclude that while high accuracy prediction of DHF is possible, care must be taken when applying DHF predictive models from one clinical setting to another.
- Open Access
- 30 October 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (10) , e0310121
MosquitoSong+: A noise-robust deep learning model for mosquito classification from wingbeat sounds
In order to assess risk of mosquito-vector borne disease and to effectively target and monitor vector control efforts, accurate information about mosquito vector population densities is needed. The traditional and still most common approach to this involves the use of traps along with manual counting and classification of mosquito species, but the costly and labor-intensive nature of this approach limits its widespread use. Numerous previous studies have sought to address this problem by developing machine learning models to automatically identify species and sex of mosquitoes based on their wingbeat sounds. Yet little work has addressed the issue of robust classification in the presence of environmental background noise, which is essential to making the approach practical. In this paper, we propose a new deep learning model, MosquitoSong+, to identify the species and sex of mosquitoes from raw wingbeat sounds so that it is robust to the environmental noise and the relative volume of the mosquito’s flight tone. The proposed model extends the existing 1D-CNN model by adjusting its architecture and introducing two data augmentation techniques during model training: noise augmentation and wingbeat volume variation. Experiments show that the new model has very good generalizability, with species classification accuracy above 80% on several wingbeat datasets with various background noise. It also has an accuracy of 93.3% for species and sex classification on wingbeat sounds overlaid with various background noises. These results suggest that the proposed approach may be a practical means to develop classification models that can perform well in the field.
- 30 October 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (10) , e0310121
Open AccessMosquitoSong+: A noise-robust deep learning model for mosquito classification from wingbeat sounds
In order to assess risk of mosquito-vector borne disease and to effectively target and monitor vector control efforts, accurate information about mosquito vector population densities is needed. The traditional and still most common approach to this involves the use of traps along with manual counting and classification of mosquito species, but the costly and labor-intensive nature of this approach limits its widespread use. Numerous previous studies have sought to address this problem by developing machine learning models to automatically identify species and sex of mosquitoes based on their wingbeat sounds. Yet little work has addressed the issue of robust classification in the presence of environmental background noise, which is essential to making the approach practical. In this paper, we propose a new deep learning model, MosquitoSong+, to identify the species and sex of mosquitoes from raw wingbeat sounds so that it is robust to the environmental noise and the relative volume of the mosquito’s flight tone. The proposed model extends the existing 1D-CNN model by adjusting its architecture and introducing two data augmentation techniques during model training: noise augmentation and wingbeat volume variation. Experiments show that the new model has very good generalizability, with species classification accuracy above 80% on several wingbeat datasets with various background noise. It also has an accuracy of 93.3% for species and sex classification on wingbeat sounds overlaid with various background noises. These results suggest that the proposed approach may be a practical means to develop classification models that can perform well in the field.
- Open Access
Machine Learning Prediction of Stroke Occurrence: A Systematic Review
Background Machine learning is nowadays commonly used for disease prediction, including cardiovascular disease. There is growing evidence of the effectiveness of machine learning algorithms for stroke risk prediction models. Aims A systematic review was conducted to identify and comprehensively evaluate the available evidence. Summary of review Relevant studies were identified from the three electronic databases (i) MEDLINE via Pubmed, (ii) Scopus, and (iii) IEEE Xplore from inception to 1st December 2020. Out of 12,626 studies identified, 40 used machine learning for ischemic or hemorrhagic stroke risk prediction models. Synthesis without meta-analysis identified that a boosting algorithm (median C-statistics = 0.9 (interquartile range [IQR]: 0.88-0.92)), and neural network (median C-statistic = 0.80 (IQR: 0.77-0.92)) performed best among ML models in the low risk of bias studies. Moreover, a boosting algorithm also performed best in overall (both low and high risk of bias) studies (median C-statistic = 0.92 (IQR: 0.90-0.95)). Conclusions The systematic review found promising results of the ML algorithm model performances compare with the gold standard conventional models, such as FSRP (C-statistic 0.653) and revised FSRP (C-statistic 0.716). In term of the algorithm, boosting and neural networks are robust, but are considered as black-box models, since they are composed of non-linearity and complex algorithms. It remains questionable whether a physician would adapt these algorithms to use in a real clinical setting. Moreover, less than half of the studies (16 out of 40) were at low risk of bias in our systematic review. More researches with good methodology and study design, alongside explainable and good performance models, may become available in the future. Trial Registration Information The International Prospective Register of Systematic Reviews (PROSPERO) database (ID: CRD42021234081).
Open AccessMachine Learning Prediction of Stroke Occurrence: A Systematic Review
Background Machine learning is nowadays commonly used for disease prediction, including cardiovascular disease. There is growing evidence of the effectiveness of machine learning algorithms for stroke risk prediction models. Aims A systematic review was conducted to identify and comprehensively evaluate the available evidence. Summary of review Relevant studies were identified from the three electronic databases (i) MEDLINE via Pubmed, (ii) Scopus, and (iii) IEEE Xplore from inception to 1st December 2020. Out of 12,626 studies identified, 40 used machine learning for ischemic or hemorrhagic stroke risk prediction models. Synthesis without meta-analysis identified that a boosting algorithm (median C-statistics = 0.9 (interquartile range [IQR]: 0.88-0.92)), and neural network (median C-statistic = 0.80 (IQR: 0.77-0.92)) performed best among ML models in the low risk of bias studies. Moreover, a boosting algorithm also performed best in overall (both low and high risk of bias) studies (median C-statistic = 0.92 (IQR: 0.90-0.95)). Conclusions The systematic review found promising results of the ML algorithm model performances compare with the gold standard conventional models, such as FSRP (C-statistic 0.653) and revised FSRP (C-statistic 0.716). In term of the algorithm, boosting and neural networks are robust, but are considered as black-box models, since they are composed of non-linearity and complex algorithms. It remains questionable whether a physician would adapt these algorithms to use in a real clinical setting. Moreover, less than half of the studies (16 out of 40) were at low risk of bias in our systematic review. More researches with good methodology and study design, alongside explainable and good performance models, may become available in the future. Trial Registration Information The International Prospective Register of Systematic Reviews (PROSPERO) database (ID: CRD42021234081).
- Open Access
- 4 October 2023
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 18 (10) , e0291389
The effect of 3D stereopsis and hand-tool alignment on learning effectiveness and skill transfer of a VR-based simulator for dental training
Recent years have seen the proliferation of VR-based dental simulators using a wide variety of different VR configurations with varying degrees of realism. Important aspects distinguishing VR hardware configurations are 3D stereoscopic rendering and visual alignment of the user’s hands with the virtual tools. New dental simulators are often evaluated without analysing the impact of these simulation aspects. In this paper, we seek to determine the impact of 3D stereoscopic rendering and of hand-tool alignment on the teaching effectiveness and skill assessment accuracy of a VR dental simulator. We developed a bimanual simulator using an HMD and two haptic devices that provides an immersive environment with both 3D stereoscopic rendering and hand-tool alignment. We then independently controlled for each of the two aspects of the simulation. We trained four groups of students in root canal access opening using the simulator and measured the virtual and real learning gains. We quantified the real learning gains by pre- and post-testing using realistic plastic teeth and the virtual learning gains by scoring the training outcomes inside the simulator. We developed a scoring metric to automatically score the training outcomes that strongly correlates with experts’ scoring of those outcomes. We found that hand-tool alignment has a positive impact on virtual and real learning gains, and improves the accuracy of skill assessment. We found that stereoscopic 3D had a negative impact on virtual and real learning gains, however it improves the accuracy of skill assessment. This finding is counter-intuitive, and we found eye-tooth distance to be a confounding variable of stereoscopic 3D, as it was significantly lower for the monoscopic 3D condition and negatively correlates with real learning gain. The results of our study provide valuable information for the future design of dental simulators, as well as simulators for other high-precision psycho-motor tasks.
- 4 October 2023
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 18 (10) , e0291389
Open AccessThe effect of 3D stereopsis and hand-tool alignment on learning effectiveness and skill transfer of a VR-based simulator for dental training
Recent years have seen the proliferation of VR-based dental simulators using a wide variety of different VR configurations with varying degrees of realism. Important aspects distinguishing VR hardware configurations are 3D stereoscopic rendering and visual alignment of the user’s hands with the virtual tools. New dental simulators are often evaluated without analysing the impact of these simulation aspects. In this paper, we seek to determine the impact of 3D stereoscopic rendering and of hand-tool alignment on the teaching effectiveness and skill assessment accuracy of a VR dental simulator. We developed a bimanual simulator using an HMD and two haptic devices that provides an immersive environment with both 3D stereoscopic rendering and hand-tool alignment. We then independently controlled for each of the two aspects of the simulation. We trained four groups of students in root canal access opening using the simulator and measured the virtual and real learning gains. We quantified the real learning gains by pre- and post-testing using realistic plastic teeth and the virtual learning gains by scoring the training outcomes inside the simulator. We developed a scoring metric to automatically score the training outcomes that strongly correlates with experts’ scoring of those outcomes. We found that hand-tool alignment has a positive impact on virtual and real learning gains, and improves the accuracy of skill assessment. We found that stereoscopic 3D had a negative impact on virtual and real learning gains, however it improves the accuracy of skill assessment. This finding is counter-intuitive, and we found eye-tooth distance to be a confounding variable of stereoscopic 3D, as it was significantly lower for the monoscopic 3D condition and negatively correlates with real learning gain. The results of our study provide valuable information for the future design of dental simulators, as well as simulators for other high-precision psycho-motor tasks.
- Open Access
- 6 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 285-290
MosquitoSongSense: IoT-based mosquito wingbeat data collection system
- 6 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 285-290
Open AccessMosquitoSongSense: IoT-based mosquito wingbeat data collection system
- 22 June 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (4) , 5189-5205
A deep learning-based pipeline for mosquito detection and classification from wingbeat sounds
- 22 June 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (4) , 5189-5205
A deep learning-based pipeline for mosquito detection and classification from wingbeat sounds
- Open Access
- 8 March 2021
- Published by Public Library of Science (PLoS) in PLOS Neglected Tropical Diseases
- Vol. 15 (3) , e0009122
Added-value of mosquito vector breeding sites from street view images in the risk mapping of dengue incidence in Thailand
Dengue is an emerging vector-borne viral disease across the world. The primary dengue mosquito vectors breed in containers with sufficient water and nutrition. Outdoor containers can be detected from geotagged images using state-of-the-art deep learning methods. In this study, we utilize such container information from street view images in developing a risk mapping model and determine the added value of including container information in predicting dengue risk. We developed seasonal-spatial models in which the target variable dengue incidence was explained using weather and container variable predictors. Linear mixed models with fixed and random effects are employed in our models to account for different characteristics of containers and weather variables. Using data from three provinces of Thailand between 2015 and 2018, the models are developed at the sub-district level resolution to facilitate the development of effective targeted intervention strategies. The performance of the models is evaluated with two baseline models: a classic linear model and a linear mixed model without container information. The performance evaluated with the correlation coefficients, R-squared, and AIC shows the proposed model with the container information outperforms both baseline models in all three provinces. Through sensitivity analysis, we investigate the containers that have a high impact on dengue risk. Our findings indicate that outdoor containers identified from street view images can be a useful data source in building effective dengue risk models and that the resulting models have potential in helping to target container elimination interventions.
- 8 March 2021
- Published by Public Library of Science (PLoS) in PLOS Neglected Tropical Diseases
- Vol. 15 (3) , e0009122
Open AccessAdded-value of mosquito vector breeding sites from street view images in the risk mapping of dengue incidence in Thailand
Dengue is an emerging vector-borne viral disease across the world. The primary dengue mosquito vectors breed in containers with sufficient water and nutrition. Outdoor containers can be detected from geotagged images using state-of-the-art deep learning methods. In this study, we utilize such container information from street view images in developing a risk mapping model and determine the added value of including container information in predicting dengue risk. We developed seasonal-spatial models in which the target variable dengue incidence was explained using weather and container variable predictors. Linear mixed models with fixed and random effects are employed in our models to account for different characteristics of containers and weather variables. Using data from three provinces of Thailand between 2015 and 2018, the models are developed at the sub-district level resolution to facilitate the development of effective targeted intervention strategies. The performance of the models is evaluated with two baseline models: a classic linear model and a linear mixed model without container information. The performance evaluated with the correlation coefficients, R-squared, and AIC shows the proposed model with the container information outperforms both baseline models in all three provinces. Through sensitivity analysis, we investigate the containers that have a high impact on dengue risk. Our findings indicate that outdoor containers identified from street view images can be a useful data source in building effective dengue risk models and that the resulting models have potential in helping to target container elimination interventions.