Song-Quan Ong

Dr. Song-Quan Ong

Universiti of Malaysia Sabah, Kota Kinabalu, Sabah, Malaysia
orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 February 2021 in Infectious Disease Reports
Verified
Journal Article
Open access
PDF file
Song-Quan OngSong-Quan Ong
Song-Quan Ong
HAHamdan Ahmad
Hamdan Ahmad
ANAhmad Mohiddin Mohd Ngesom
Ahmad Mohiddin Mohd Ngesom
We aim to investigate the effect of large-scale human movement restrictions during the COVID-19 lockdown on both the dengue transmission and vector occurrences. This study compared the weekly dengue incidences during the period of lockdown to the previous years (2015 to 2019) and a Seasonal Autoregressive Integrated Moving Average (SARIMA) model that expected no movement restrictions. We found that the trend of dengue incidence during the first two weeks (stage 1) of lockdown decreased significantly with the incidences lower than the lower confidence level (LCL) of SARIMA. By comparing the magnitude of the gradient of decrease, the trend is 319% steeper than the trend observed in previous years and 650% steeper than the simulated model, indicating that the control of population movement did reduce dengue transmission. However, starting from stage 2 of lockdown, the dengue incidences demonstrated an elevation and earlier rebound by four weeks and grew with an exponential pattern. We revealed that Aedes albopictus is the predominant species and demonstrated a strong correlation with the locally reported dengue incidences, and therefore we proposed the possible diffusive effect of the vector that led to a higher acceleration of incidence rate.

10 May 2021 in Scientific Reports
Verified
Journal Article
Open access
PDF file
Song-Quan OngSong-Quan Ong
Song-Quan Ong
HAHamdan Ahmad
Hamdan Ahmad
GNGomesh Nair
Gomesh Nair
PIPradeep Isawasan
Pradeep Isawasan
AMAbdul Hafiz Ab Majid
Abdul Hafiz Ab Majid
Classification of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse) by humans remains challenging. We proposed a highly accessible method to develop a deep learning (DL) model and implement the model for mosquito image classification by using hardware that could regulate the development process. In particular, we constructed a dataset with 4120 images of Aedes mosquitoes that were older than 12 days old and had common morphological features that disappeared, and we illustrated how to set up supervised deep convolutional neural networks (DCNNs) with hyperparameter adjustment. The model application was first conducted by deploying the model externally in real time on three different generations of mosquitoes, and the accuracy was compared with human expert performance. Our results showed that both the learning rate and epochs significantly affected the accuracy, and the best-performing hyperparameters achieved an accuracy of more than 98% at classifying mosquitoes, which showed no significant difference from human-level performance. We demonstrated the feasibility of the method to construct a model with the DCNN when deployed externally on mosquitoes in real time.

5 November 2023 in Scientific Reports
Verified
Journal Article
Open access
PDF file
Song-Quan OngSong-Quan Ong
Song-Quan Ong
PIPradeep Isawasan
Pradeep Isawasan
ANAhmad Mohiddin Mohd Ngesom
Ahmad Mohiddin Mohd Ngesom
HSHanipah Shahar
Hanipah Shahar
ALAs’mAlia Md Lasim
As’mAlia Md Lasim
GNGomesh Nair
Gomesh Nair
Machine learning algorithms (ML) are receiving a lot of attention in the development of predictive models for monitoring dengue transmission rates. Previous work has focused only on specific weather variables and algorithms, and there is still a need for a model that uses more variables and algorithms that have higher performance. In this study, we use vector indices and meteorological data as predictors to develop the ML models. We trained and validated seven ML algorithms, including an ensemble ML method, and compared their performance using the receiver operating characteristic (ROC) with the area under the curve (AUC), accuracy and F1 score. Our results show that an ensemble ML such as XG Boost, AdaBoost and Random Forest perform better than the logistics regression, Naïve Bayens, decision tree, and support vector machine (SVM), with XGBoost having the highest AUC, accuracy and F1 score. Analysis of the importance of the variables showed that the container index was the least important. By removing this variable, the ML models improved their performance by at least 6% in AUC and F1 score. Our result provides a framework for future studies on the use of predictive models in the development of an early warning system.

30 December 2022 in PLOS ONE
Verified
Journal Article
Open access
PDF file
Song-Quan OngSong-Quan Ong
Song-Quan Ong
SHSuhaila Ab. Hamid
Suhaila Ab. Hamid
Insect taxonomy lies at the heart of many aspects of ecology, and identification tasks are challenging due to the enormous inter- and intraspecies variation of insects. Conventional methods used to study insect taxonomy are often tedious, time-consuming, labor intensive, and expensive, and recently, computer vision with deep learning algorithms has offered an alternative way to identify and classify insect images into their taxonomic levels. We designed the classification task according to the taxonomic ranks of insects—order, family, and genus—and compared the generalization of four state-of-the-art deep convolutional neural network (DCNN) architectures. The results show that different taxonomic ranks require different deep learning (DL) algorithms to generate high-performance models, which indicates that the design of an automated systematic classification pipeline requires the integration of different algorithms. The InceptionV3 model has advantages over other models due to its high performance in distinguishing insect order and family, which is having F1-score of 0.75 and 0.79, respectively. Referring to the performance per class, Hemiptera (order), Rhiniidae (family), and Lucilia (genus) had the lowest performance, and we discuss the possible rationale and suggest future works to improve the generalization of a DL model for taxonomic rank classification.

1 January 2015 in Parasites & Vectors
Verified
Journal Article
Open access
PDF file
Song-Quan OngSong-Quan Ong
Song-Quan Ong
ZJZairi Jaal
Zairi Jaal
The trend in chemical insecticide development has focused on improving the efficacy against mosquitoes while reducing the environmental impact. Lethal lures apply an “attract-and-kill” strategy that draws the insect to the killing agent rather than bringing the killing agent to the insect. In this study, the mosquito oviposition pheromone was extracted from the eggs of Aedes aegypti (L.) and further investigated with a combination of pheromone and granular temephos as a lethal lure. The compound caproic acid attracted significantly more egg-laying mosquitos at 1 ppm (660.83 ± 91.61) than the control (343.83 ± 56.24), which consisted of solvent only (Oviposition Activity Index: 0.316). Further investigation of the combination of caproic acid with granular temephos as a lethal lure attracted significantly more gravid female Ae. aegypti to oviposit their eggs than the temephos treated water and control. This indicated the ability of caproic acid in acting as an attractant and counters the repellency effect of temephos. Additionally, the presence of temephos in the lethal lure also restricted the hatching of the eggs and killed any larvae that hatched.

14 June 2023 in Ecology and Evolution
Verified
Journal Article
Open access
Song-Quan OngSong-Quan Ong
Song-Quan Ong
NJNurzatil Sharleeza Mat Jalaluddin
Nurzatil Sharleeza Mat Jalaluddin
KYKien Thai Yong
Kien Thai Yong
SOSu Ping Ong
Su Ping Ong
KLKooi Fong Lim
Kooi Fong Lim
SASuhaila Azhar
Suhaila Azhar
Natural history museum collections are the most important sources of information on the present and past biodiversity of our planet. Most of the information is primarily stored in analogue form, and digitization of the collections can provide further open access to the images and specimen data to address the many global challenges. However, many museums do not digitize their collections because of constraints on budgets, human resources, and technologies. To encourage the digitization process, we present a guideline that offers low‐cost and technical knowledge solutions yet balances the quality of the work and outcomes. The guideline describes three phases of digitization, namely preproduction, production, and postproduction. The preproduction phase includes human resource planning and selecting the highest priority collections for digitization. In the preproduction phase, a worksheet is provided for the digitizer to document the metadata, as well as a list of equipment needed to set up a digitizer station to image the specimens and associated labels. In the production phase, we place special emphasis on the light and color calibrations, as well as the guidelines for ISO/shutter speed/aperture to ensure a satisfactory quality of the digitized output. Once the specimen and labels have been imaged in the production phase, we demonstrate an end‐to‐end pipeline that uses optical character recognition (OCR) to transfer the physical text on the labels into a digital form and document it in a worksheet cell. A nationwide capacity workshop is then conducted to impart the guideline, and pre‐ and postcourse surveys were conducted to assess the confidence and skills acquired by the participants. This paper also discusses the challenges and future work that need to be taken forward for proper digital biodiversity data management.