Xavier Ngu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xavier Ngu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Audrey Huong
Universiti Tun Hussein Onn Malaysia
7 shared publications - Kim Gaik Tay
Universiti Tun Hussein Onn Malaysia
3 shared publications - Kok Beng Gan
Department of Electrical, Electronic & Systems Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, UKM Bangi, Malaysia
3 shared publications Lik Suong Ding
University Tun Hussein Onn Malaysia
2 shared publications- Ser Lee Loh
Universiti Teknikal Malaysia Melaka
2 shared publications - Hui Ling Chua
Industrial Polymer Sdn. Bhd
1 shared publication
Latest Publications
- 3 February 2026
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 71, 95-108
An Automatic PSO-Optimized Spatial-Temporal Neural Network for Blood Perfusion Analysis
Existing technologies for evaluating oxygen delivery efficiency and blood perfusion are limited to complex equipment, whilst accurate diagnosis of compromised microvascular circulation relies on domain-specific mathematical models. Although an ultrasonic flaw detector can be an economical alternative, developing an optimal prediction model for a specific imaging task requires manual fine-tuning and customization. This study aims to introduce an optimization-empowered hybrid spatial-temporal network design framework for rapid and nondestructive blood perfusion status classification using a simple photoacoustic system. To examine this framework, this paper conducted experiments on seventeen volunteers. External intervention was performed to modify the perfusion of the chosen body part; the oxygenation-related photoacoustic (PA) signal was recorded after the chosen site was illuminated with temporally modulated light beams of wavelengths 500 nm and 550 nm. Meanwhile, the network design and training processes used the Particle Swarm Optimization (PSO) method for efficient model convergence on the small PA dataset. Experimental results comparing the designed hybrid model against two benchmark networks trained for the same microcirculatory perfusion disturbance problem confirmed the competitiveness of the proposed model with a superior mean classification accuracy, specificity, precision, and F 1 score of 80.8 %, 92.3 %, 89.9 %, and 79.4 %, respectively. This is further supported by the result of the Receiver operating characteristic (ROC) curve, showing a good area under the curve (AUC) of 0.91. These findings offered evidence of hemodynamic fluctuations at the microcirculatory level to compensate for low oxygen under ischemic-hypoxic conditions. This study concluded that the future of this system includes its industrial and medical applications for nondestructive inspections, home use, or bedside microcirculation monitoring to optimize wound and postoperative outcomes.
- 3 February 2026
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 71, 95-108
An Automatic PSO-Optimized Spatial-Temporal Neural Network for Blood Perfusion Analysis
Existing technologies for evaluating oxygen delivery efficiency and blood perfusion are limited to complex equipment, whilst accurate diagnosis of compromised microvascular circulation relies on domain-specific mathematical models. Although an ultrasonic flaw detector can be an economical alternative, developing an optimal prediction model for a specific imaging task requires manual fine-tuning and customization. This study aims to introduce an optimization-empowered hybrid spatial-temporal network design framework for rapid and nondestructive blood perfusion status classification using a simple photoacoustic system. To examine this framework, this paper conducted experiments on seventeen volunteers. External intervention was performed to modify the perfusion of the chosen body part; the oxygenation-related photoacoustic (PA) signal was recorded after the chosen site was illuminated with temporally modulated light beams of wavelengths 500 nm and 550 nm. Meanwhile, the network design and training processes used the Particle Swarm Optimization (PSO) method for efficient model convergence on the small PA dataset. Experimental results comparing the designed hybrid model against two benchmark networks trained for the same microcirculatory perfusion disturbance problem confirmed the competitiveness of the proposed model with a superior mean classification accuracy, specificity, precision, and F 1 score of 80.8 %, 92.3 %, 89.9 %, and 79.4 %, respectively. This is further supported by the result of the Receiver operating characteristic (ROC) curve, showing a good area under the curve (AUC) of 0.91. These findings offered evidence of hemodynamic fluctuations at the microcirculatory level to compensate for low oxygen under ischemic-hypoxic conditions. This study concluded that the future of this system includes its industrial and medical applications for nondestructive inspections, home use, or bedside microcirculation monitoring to optimize wound and postoperative outcomes.
- Open Access
- 10 December 2025
- Published by FOREX Publication in International Journal of Electrical and Electronics Research
- Vol. 13 (4) , 735-748
Cascaded Segmentation-Classification Framework Optimized for Sports Action Recognition Using Still Images: A Comparative Analysis of PSO and Grid Search
The conventional method of evaluating human actions and activities requires integrating complex hardware and interpretation systems. This article proposes using a hybrid segmentation-classification system to recognize human sports action automatically based on still frames. This study tested the proposed framework on a public dataset containing images of humans performing three athletic actions: dancing, performing martial arts, and playing net sports. This framework combined a two-stage U-Net and EfficientNet-B0, and GoogleNet, and is optimized using particle swarm optimization (PSO) and grid search for cascaded segmentation and classification problems. Results indicated that PSO improves segmentation and classification accuracies by 5 % and 60 %, respectively, compared to the conventional grid search. The PSO segmentation results showed good agreement between the predicted mask and its ground truth, with overlapping scores of 0.85-0.9. The consecutive classification experiments revealed a slight superiority in the performance of EfficientNet-B0 with an accuracy of 0.86-0.88 over GoogleNet (~0.82-0.84), which also showed a lower convergence efficiency. While no significant difference was observed in the recall scores of EfficientNet between the weighted and unweighted loss methods, GoogleNet showed a considerable improvement in the true positive rates from 0.6177 to 0.7160 using the weighted loss strategy. Despite using the weighted loss method, there were negligible improvements in the performance of grid search-optimized networks. The poor classification results suggest low model generalization ability using manual tuning. This study concluded that the PSO-optimized cascaded segmentation-classification framework could potentially leverage advancements in human movement evaluation and rehabilitation assessment for sports applications. This system can be improved by adopting larger datasets or collaborative machine learning training to enhance model convergence for practical rehabilitation applications to assess users’ physical and fitness performance.
- 10 December 2025
- Published by FOREX Publication in International Journal of Electrical and Electronics Research
- Vol. 13 (4) , 735-748
Open AccessCascaded Segmentation-Classification Framework Optimized for Sports Action Recognition Using Still Images: A Comparative Analysis of PSO and Grid Search
The conventional method of evaluating human actions and activities requires integrating complex hardware and interpretation systems. This article proposes using a hybrid segmentation-classification system to recognize human sports action automatically based on still frames. This study tested the proposed framework on a public dataset containing images of humans performing three athletic actions: dancing, performing martial arts, and playing net sports. This framework combined a two-stage U-Net and EfficientNet-B0, and GoogleNet, and is optimized using particle swarm optimization (PSO) and grid search for cascaded segmentation and classification problems. Results indicated that PSO improves segmentation and classification accuracies by 5 % and 60 %, respectively, compared to the conventional grid search. The PSO segmentation results showed good agreement between the predicted mask and its ground truth, with overlapping scores of 0.85-0.9. The consecutive classification experiments revealed a slight superiority in the performance of EfficientNet-B0 with an accuracy of 0.86-0.88 over GoogleNet (~0.82-0.84), which also showed a lower convergence efficiency. While no significant difference was observed in the recall scores of EfficientNet between the weighted and unweighted loss methods, GoogleNet showed a considerable improvement in the true positive rates from 0.6177 to 0.7160 using the weighted loss strategy. Despite using the weighted loss method, there were negligible improvements in the performance of grid search-optimized networks. The poor classification results suggest low model generalization ability using manual tuning. This study concluded that the PSO-optimized cascaded segmentation-classification framework could potentially leverage advancements in human movement evaluation and rehabilitation assessment for sports applications. This system can be improved by adopting larger datasets or collaborative machine learning training to enhance model convergence for practical rehabilitation applications to assess users’ physical and fitness performance.
- Open Access
- 16 July 2024
- Published by Universiti Putra Malaysia in Pertanika Journal of Science and Technology
- Vol. 32 (4) , 1461-1479
Blood CO Status Classification Using UV-VIS Spectroscopy and PSO-optimized 1D-CNN Model
Rapid and effective blood carbon monoxide (CO) assessment is of great importance, especially in estimating CO-related morbidity and instituting effective preventive measures. The conventional detection methods using CO breath analysis lack sensitivity, while collecting biological fluid samples for CO level measurement is prone to external contamination and expensive for frequent use. This study proposes a one-dimensional convolutional neural network (1D-CNN) consisting of three stacked biconvolutional layers for binary classification of blood CO status using the diffuse reflectance spectroscopy technique. Iterative particle swarm optimization (PSO) has efficiently found the best network parameters to learn important features from the reflectance spectroscopy data. The findings showed good testing accuracy, specificity, and precision of 92.9%, 90%, and 89.7%, respectively, and a high sensitivity of 96.3% in determining abnormal blood CO among smokers using the proposed CNN network. Comparisons with eight existing machine learning and deep learning models revealed the proposed method’s effectiveness in classifying blood CO status while reducing computing time by 8–13 folds. The findings of this work provide new insights that are valuable for researchers in neural network design automation, healthcare management, and skin-related research, specifically for application in nondestructive evaluation and clinical decision-making.
- 16 July 2024
- Published by Universiti Putra Malaysia in Pertanika Journal of Science and Technology
- Vol. 32 (4) , 1461-1479
Open AccessBlood CO Status Classification Using UV-VIS Spectroscopy and PSO-optimized 1D-CNN Model
Rapid and effective blood carbon monoxide (CO) assessment is of great importance, especially in estimating CO-related morbidity and instituting effective preventive measures. The conventional detection methods using CO breath analysis lack sensitivity, while collecting biological fluid samples for CO level measurement is prone to external contamination and expensive for frequent use. This study proposes a one-dimensional convolutional neural network (1D-CNN) consisting of three stacked biconvolutional layers for binary classification of blood CO status using the diffuse reflectance spectroscopy technique. Iterative particle swarm optimization (PSO) has efficiently found the best network parameters to learn important features from the reflectance spectroscopy data. The findings showed good testing accuracy, specificity, and precision of 92.9%, 90%, and 89.7%, respectively, and a high sensitivity of 96.3% in determining abnormal blood CO among smokers using the proposed CNN network. Comparisons with eight existing machine learning and deep learning models revealed the proposed method’s effectiveness in classifying blood CO status while reducing computing time by 8–13 folds. The findings of this work provide new insights that are valuable for researchers in neural network design automation, healthcare management, and skin-related research, specifically for application in nondestructive evaluation and clinical decision-making.
- Open Access
- 31 October 2021
- Published by The Korean Society of Medical Informatics in Healthcare Informatics Research
- Vol. 27 (4) , 298-306
Five-Class Classification of Cervical Pap Smear Images: A Study of CNN-Error-Correcting SVM Models
Objectives: Different complex strategies of fusing handcrafted descriptors and features from convolutional neural network (CNN) models have been studied, mainly for two-class Papanicolaou (Pap) smear image classification. This paper explores a simplified system using combined binary coding for a five-class version of this problem.Methods: This system extracted features from transfer learning of AlexNet, VGG19, and ResNet50 networks before reducing this problem into multiple binary sub-problems using error-correcting coding. The learners were trained using the support vector machine (SVM) method. The outputs of these classifiers were combined and compared to the true class codes for the final prediction.Results: Despite the superior performance of VGG19-SVM, with mean ± standard deviation accuracy and sensitivity of 80.68% ± 2.00% and 80.86% ± 0.45%, respectively, this model required a long training time. There were also false-negative cases using both the VGGNet-SVM and ResNet-SVM models. AlexNet-SVM was more efficient in terms of running speed and prediction consistency. Our findings also showed good diagnostic ability, with an area under the curve of approximately 0.95. Further investigation also showed good agreement between our research outcomes and that of the state-of-the-art methods, with specificity ranging from 93% to 100%.Conclusions: We believe that the AlexNet-SVM model can be conveniently applied for clinical use. Further research could include the implementation of an optimization algorithm for hyperparameter tuning, as well as an appropriate selection of experimental design to improve the efficiency of Pap smear image classification.
- 31 October 2021
- Published by The Korean Society of Medical Informatics in Healthcare Informatics Research
- Vol. 27 (4) , 298-306
Open AccessFive-Class Classification of Cervical Pap Smear Images: A Study of CNN-Error-Correcting SVM Models
Objectives: Different complex strategies of fusing handcrafted descriptors and features from convolutional neural network (CNN) models have been studied, mainly for two-class Papanicolaou (Pap) smear image classification. This paper explores a simplified system using combined binary coding for a five-class version of this problem.Methods: This system extracted features from transfer learning of AlexNet, VGG19, and ResNet50 networks before reducing this problem into multiple binary sub-problems using error-correcting coding. The learners were trained using the support vector machine (SVM) method. The outputs of these classifiers were combined and compared to the true class codes for the final prediction.Results: Despite the superior performance of VGG19-SVM, with mean ± standard deviation accuracy and sensitivity of 80.68% ± 2.00% and 80.86% ± 0.45%, respectively, this model required a long training time. There were also false-negative cases using both the VGGNet-SVM and ResNet-SVM models. AlexNet-SVM was more efficient in terms of running speed and prediction consistency. Our findings also showed good diagnostic ability, with an area under the curve of approximately 0.95. Further investigation also showed good agreement between our research outcomes and that of the state-of-the-art methods, with specificity ranging from 93% to 100%.Conclusions: We believe that the AlexNet-SVM model can be conveniently applied for clinical use. Further research could include the implementation of an optimization algorithm for hyperparameter tuning, as well as an appropriate selection of experimental design to improve the efficiency of Pap smear image classification.
- 14 April 2021
- Published by Springer Nature
- p. 61-71
Integrated Circuit Characterisation Using Low-Cost Coaxial Probe Measurement
- 14 April 2021
- Published by Springer Nature
- p. 61-71
Integrated Circuit Characterisation Using Low-Cost Coaxial Probe Measurement
- 14 April 2021
- Published by Springer Nature
- p. 73-85
A More Efficient and Accurate Approach to Characterise an IC into Macro Model
- 14 April 2021
- Published by Springer Nature
- p. 73-85
A More Efficient and Accurate Approach to Characterise an IC into Macro Model
- Open Access
- 1 August 2020
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 19 (2) , 793-801
Three-wavelength system for practical application in skin oximetry: simultaneous equations with prediction-correction approach
This paper presented the use of a three-wavelength system coupled with a prediction-correction model for the measurement of a person’s tissue oxygen levels and in the efforts towards the development of a field-portable system. This study considered light wavelength of market-available emitters in the range 500 − 650 nm for its practical implementation. This approach required the use of light attenuation and hemoglobin absorptivity information of three different wavelengths in determining tissue oxygen saturation value, StO2. It was found through the analysis of results using Monte Carlo method that considerable improvement in the accuracy of the predictions was obtained using the corrective models (ρ =0.874). The low mean prediction errors of similar magnitude, not exceeding 4 %, given by two wavelength combinations 538, 560, 633 nm and 538, 560, 650 nm were observed for signals with signal-noise ratio (SNR) of down to 30 dB. A significant statistical difference was found between the prediction errors and the wavelength combination used under this noise condition (ρ =0.011). This work concluded that the findings of this study provide insights into technology implementation of skin oximetry and the possible impacts it might have in medical arena.
- 1 August 2020
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 19 (2) , 793-801
Open AccessThree-wavelength system for practical application in skin oximetry: simultaneous equations with prediction-correction approach
This paper presented the use of a three-wavelength system coupled with a prediction-correction model for the measurement of a person’s tissue oxygen levels and in the efforts towards the development of a field-portable system. This study considered light wavelength of market-available emitters in the range 500 − 650 nm for its practical implementation. This approach required the use of light attenuation and hemoglobin absorptivity information of three different wavelengths in determining tissue oxygen saturation value, StO2. It was found through the analysis of results using Monte Carlo method that considerable improvement in the accuracy of the predictions was obtained using the corrective models (ρ =0.874). The low mean prediction errors of similar magnitude, not exceeding 4 %, given by two wavelength combinations 538, 560, 633 nm and 538, 560, 650 nm were observed for signals with signal-noise ratio (SNR) of down to 30 dB. A significant statistical difference was found between the prediction errors and the wavelength combination used under this noise condition (ρ =0.011). This work concluded that the findings of this study provide insights into technology implementation of skin oximetry and the possible impacts it might have in medical arena.
- 1 January 2019
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 42, 59-66
Portable and Rapid In Vivo Imaging of Tissue Oxygenation Changes Induced by Skin Perfusion Pressure
Current imaging systems available in the study of tissue hemodynamics and vascular reactivity are typically complex and bulky, hence limiting their applications to laboratory use. The aim of this study is to present the dynamics of skin oxygen level with changes in the microcirculatory perfusion monitored using a developed field portable, handheld tri-wavelength imaging system. The skin oxygen saturation level (StO2) t was measured in-vivo in a span of 12 minutes in the ventral forearm of seven healthy humans at rest, before and after supra-diastolic and supra-systolic pressure inflations. The findings of this work showed statistical significance in the difference between the mean StO2 values in baseline and that following ischemic episodes with rho <= 0.03. The values returned to baseline, although of different magnitude for each individual, within 2 min (rho =0.217) during reperfusion could suggest the time required for the resume of normal autoregulation mechanisms and vasomotion reactivities in the recruits. This study concluded that the developed imaging system could find potential application in self-healthcare management and may help to improve patient care in remote or rural locations.
- 1 January 2019
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 42, 59-66
Portable and Rapid In Vivo Imaging of Tissue Oxygenation Changes Induced by Skin Perfusion Pressure
Current imaging systems available in the study of tissue hemodynamics and vascular reactivity are typically complex and bulky, hence limiting their applications to laboratory use. The aim of this study is to present the dynamics of skin oxygen level with changes in the microcirculatory perfusion monitored using a developed field portable, handheld tri-wavelength imaging system. The skin oxygen saturation level (StO2) t was measured in-vivo in a span of 12 minutes in the ventral forearm of seven healthy humans at rest, before and after supra-diastolic and supra-systolic pressure inflations. The findings of this work showed statistical significance in the difference between the mean StO2 values in baseline and that following ischemic episodes with rho <= 0.03. The values returned to baseline, although of different magnitude for each individual, within 2 min (rho =0.217) during reperfusion could suggest the time required for the resume of normal autoregulation mechanisms and vasomotion reactivities in the recruits. This study concluded that the developed imaging system could find potential application in self-healthcare management and may help to improve patient care in remote or rural locations.
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 9-12
Multispectral imaging system for clinical assessment of superficial wound tissue oxygenation
Sufficient tissue oxygenation is a factor that warrants fast progression of wound healing as the latter would be impeded under the condition of inappropriate oxygen supply to wound tissues. This work aims to investigate average wound tissue oxygenation value during wound healing process. We used Extended Modified Lambert Beer model to predict tissue oxygen saturation (StO2) based on spectroscopic data collected from multispectral imaging system. We recruited two healthy Asians with acute wound at different skin sites. Multispectral data were collected at each of their scheduled clinical visit starting from the commencement of wound treatment until the wound dressing is no longer required as evaluated by medical professionals. The quantification work is based on a priori knowledge of absorption coefficients of blood hemoglobin components in the wavelength range of 520-600 nm. This study observed an increase in mean StO2 value to a maximum value of 71.73±2.35% as micro-angiogenesis takes place. This value is gradually decreased to 61.87±3.54% as skin keratinized. We concluded that multispectral imaging technique and the employed quantification strategy is potentially useful as a means for noncontact and noninvasive monitoring of wound healing process, and this system may be suitable for routine clinical use.
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 9-12
Multispectral imaging system for clinical assessment of superficial wound tissue oxygenation
Sufficient tissue oxygenation is a factor that warrants fast progression of wound healing as the latter would be impeded under the condition of inappropriate oxygen supply to wound tissues. This work aims to investigate average wound tissue oxygenation value during wound healing process. We used Extended Modified Lambert Beer model to predict tissue oxygen saturation (StO2) based on spectroscopic data collected from multispectral imaging system. We recruited two healthy Asians with acute wound at different skin sites. Multispectral data were collected at each of their scheduled clinical visit starting from the commencement of wound treatment until the wound dressing is no longer required as evaluated by medical professionals. The quantification work is based on a priori knowledge of absorption coefficients of blood hemoglobin components in the wavelength range of 520-600 nm. This study observed an increase in mean StO2 value to a maximum value of 71.73±2.35% as micro-angiogenesis takes place. This value is gradually decreased to 61.87±3.54% as skin keratinized. We concluded that multispectral imaging technique and the employed quantification strategy is potentially useful as a means for noncontact and noninvasive monitoring of wound healing process, and this system may be suitable for routine clinical use.