2005 - 2009 Doctor of Philosophy, School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore
2001 - 2005 Bachelor of Mechanical Engineering, Faculty of Mechanical Engineering, Universiti Teknologi Malaysia
RESEARCH, TEACHING, or OTHER INTERESTS
Mechanical Engineering, Multidisciplinary, Modeling and Simulation, Numerical Analysis
117
Scopus Publications
Scopus Publications
Mushroom growth stage detection and maturity prediction using lightweight vision and logistic models for smart cultivation Wanpeng Fan, Yeong Shiong Chiew, Ean Hin Ooi, Xin Wang Smart Agricultural Technology, 2026 Accurate growth stage detection and prediction of full-maturity time are critical for optimising yield, scheduling harvests and improve labour efficiency in mushroom cultivation. However, existing vision-based methods mainly address stage classification without addressing temporal prediction. To overcome this limitation, this study proposes a lightweight detection–prediction framework (YOLO-Pulmonarius, YOLO-P) specifically designed for oyster mushrooms. A Swin Transformer v2 module was integrated into the YOLOv12 detector to improve long-range feature modelling and enhance robustness against glare, occlusion, and overlapping fruiting bodies. Temporal prediction of full maturity was achieved through logistic curve fitting coupled with recursive Gaussian fusion, incorporating area normalisation and dynamic priors for adaptive temporal modelling. The proposed framework was trained on 3,867 annotated images across three growth stages of grey oyster mushrooms. The proposed YOLO-P detection model achieved an average mAP@50 of 97.2%, with the highest performance of 98.6% in the Expired stage. Logistic modelling revealed a stable growth inflection at 62 h (≈2.6 days). The full-maturity time prediction achieved a mean absolute error of 0.42 days and a Pearson correlation of 0.86, outperforming ablation variants without area normalisation (0.49 days, R = 0.79) and without dynamic priors (0.52 days, R = 0.82). The proposed framework establishes a closed-loop pipeline that integrates stage detection with temporal prediction. Its lightweight architecture enables deployment on edge devices, offering a practical solution for real-time monitoring and automated scheduling in smart mushroom cultivation.
Convection heat transfer enhancement of a vertically heated surface via rising bubbles in graphene oxide (GO) nanofluid Li Teng Siow, Jun Rong Lee, Atsuhide Kitagawa, Ean Hin Ooi, Ee Von Lau International Journal of Heat and Mass Transfer, 2025 • Bubbles rising in GO-water nanofluid capable of enhancing heat transfer. • 0.1 g/L GO-water nanofluid has overall better heat transfer performance. • Heat transfer enhanced with increased air injection rate from 6 to 36 mL/min. • Optimum GO concentration exists for better heat transfer performance. • Maximum enhancement of 145 % was achieved by 0.2 g/L GO-water nanofluid at 36 mL/min. Nanofluids enhance heat transfer due to the addition of highly thermally conductive nanoparticles into the base fluid, which has low thermal conductivity. On the other hand, rising gas bubbles in a liquid medium can enhance heat transfer over a heated surface due to mixing effects induced by the vortices caused by the bubbles rising motion. This study aims to integrate the effects of graphene oxide (GO) nanofluids and rising air bubbles to enhance heat transfer over a vertically heated surface. Air bubbles were injected into deionised water-based GO nanofluids, and heat transfer performance was quantified by the heat transfer coefficient ( h ¯ ). With bubbles rising in a GO-water nanofluid of concentration 0.1 g/L and an air injection rate of 6 mL/min, h ¯ increased by 22.3 % compared to deionised water without bubbles. However, when nanofluids were used without bubble injection, the heat transfer performance deteriorated, proving the effectiveness of rising bubbles in enhancing heat transfer in GO-water nanofluids. The improvement was higher for rising bubbles in GO-water nanofluids than bubbles in deionised water alone. Moreover, the heat transfer coefficient increased with increasing air injection rates. Among all the liquid mediums tested, the 0.1 g/L GO-water nanofluid outperformed deionised water, 0.05 g/L GO and 0.2 g/L GO, achieving the highest h ¯ value of 1044 W/(m 2 ·K) at an air injection rate of 36 mL/min. Therefore, with optimum concentration, rising air bubbles in GO-water nanofluid are capable of dissipating heat and enhancing heat transfer on a heated surface.
Amogel: a multi-omics classification framework using associative graph neural networks with prior knowledge for biomarker identification Chia Yan Tan, Huey Fang Ong, Chern Hong Lim, Mei Sze Tan, Ean Hin Ooi, KokSheik Wong BMC Bioinformatics, 2025 The advent of high-throughput sequencing technologies, such as DNA microarray and DNA sequencing, has enabled effective analysis of cancer subtypes and targeted treatment. Furthermore, numerous studies have highlighted the capability of graph neural networks (GNN) to model complex biological systems and capture non-linear interactions in high-throughput data. GNN has proven to be useful in leveraging multiple types of omics data, including prior biological knowledge from various sources, such as transcriptomics, genomics, proteomics, and metabolomics, to improve cancer classification. However, current works do not fully utilize the non-linear learning potential of GNN and lack of the integration ability to analyse high-throughput multi-omics data simultaneously with prior biological knowledge. Nevertheless, relying on limited prior knowledge in generating gene graphs might lead to less accurate classification due to undiscovered significant gene-gene interactions, which may require expert intervention and can be time-consuming. Hence, this study proposes a graph classification model called associative multi-omics graph embedding learning (AMOGEL) to effectively integrate multi-omics datasets and prior knowledge through GNN coupled with association rule mining (ARM). AMOGEL employs an early fusion technique using ARM to mine intra-omics and inter-omics relationships, forming a multi-omics synthetic information graph before the model training. Moreover, AMOGEL introduces multi-dimensional edges, with multi-omics gene associations or edges as the main contributors and prior knowledge edges as auxiliary contributors. Additionally, it uses a gene ranking technique based on attention scores, considering the relationships between neighbouring genes. Several experiments were performed on BRCA and KIPAN cancer subtypes to demonstrate the integration of multi-omics datasets (miRNA, mRNA, and DNA methylation) with prior biological knowledge of protein-protein interactions, KEGG pathways and Gene Ontology. The experimental results showed that the AMOGEL outperformed the current state-of-the-art models in terms of classification accuracy, F1 score and AUC score. The findings of this study represent a crucial step forward in advancing the effective integration of multi-omics data and prior knowledge to improve cancer subtype classification.
Effects of skin surface roughness on the passive and active thermographic detection of melanoma: A numerical analysis Hou Y. Mok, Ean H. Ooi, Yeong S. Chiew, N. Pamidi, Ean T. Ooi International Journal of Heat and Mass Transfer, 2025 Infrared thermography, while promising as a non-invasive melanoma detection has not been adopted clinically. This is because the thermal signal induced by early stage melanoma is of the same magnitude as those induced by skin surface roughness. This causes the apparent skin temperature variation that can diminish the tumour signal. To investigate this, computational models of the human skin with four different surface roughness, R a = 0, 5.0 ± 0.1, 10.3 ± 0.2 and 15.6 ± 0 . 3 μ m , were constructed. Heat transfer across the skin was described using bioheat transfer. Simulations were carried out for passive thermography (PT) and dynamic thermal imaging (DT). Numerical results indicated that both PT and DT were capable of detecting the presence of T1 to T4 melanoma if the skin surface roughness within the field-of-view (FOV) of the thermal camera is uniform. However, if differences in surface roughness exist within the FOV, the roughness-induced thermal fluctuations were 2 and 1.33 times larger than those induced by T1 and T2 melanoma, respectively. With DT, the parameters quantifying the difference in the thermal recovery curves (TRCs) between two healthy regions of interest with different surface roughness were significantly greater than those caused by the presence of T1 and T2 melanoma. The results suggest that skin surface roughness can diminish the thermal signal from T1 and T2 melanoma during infrared thermography, leading to false-negative detection. Future studies should explore incorporating surface roughness identification as part of the protocol for thermographic melanoma detection.
Stochastic virtual patient-guided mechanical ventilation treatment: A virtual patient study with mechanical power consideration Christopher Yew Shuen Ang, Yeong Shiong Chiew, Xin Wang, Ean Hin Ooi, Mohd Basri Mat Nor, Matthew E. Cove, Cong Zhou, J. Geoffrey Chase IFAC Journal of Systems and Control, 2025 Background and Objective: Computerised decision support systems (CDSS) in mechanical ventilation (MV) provide individualised, closed-loop treatment but often require extensive input parameters, which are challenging to obtain continuously in clinical settings. Many also fail to incorporate mechanical power ( MP ) and MP ratio — recently identified as significant predictors of patient outcomes. This study introduces the Stochastic Virtual Patient Ventilation Protocol (SVP VENT), a model-based CDSS addressing these limitations. Methods: The SVP VENT Protocol integrates a stochastic virtual patient model to predict temporal lung elastance, E r s , trends and deliver closed-loop, lung protective ventilation minimising MP ratio and driving pressure. The protocol was validated against the VENT and SiVENT protocols using an established virtual patient platform comprising over 1229 h of both volume control (VC) and pressure control (PC) retrospective MV data. Patient responses were monitored to ensure adherence to accepted clinical safety guidelines. Results: The SVP VENT protocol consistently outperformed retrospective clinical data, VENT and SiVENT protocols in ensuring adherence to clinical safety metrics, achieving an all-adherence rate of ∼ 57% and ∼ 67% for the VC and PC cohorts, respectively. Across cohorts, the protocol maintained MP and MP ratio levels below safety thresholds (12 J/min and 4.5, respectively), and extended intervention intervals up to 3 h, potentially reducing clinical workload. Conclusion: Overall, the virtual trial demonstrates the SVP VENT protocol’s potential to enhance MV management by extending intervention intervals, while maintaining patient safety. These findings support initial clinical trials to evaluate the protocol’s impact on clinical workload and patient safety over prolonged monitoring periods, facilitating its integration into standard clinical practices.
Lattice Boltzmann-based microchannel concentration mixing with surface roughness-mediated flow dynamics Lit Kean Chai, Chin Vern Yeoh, Ean Hin Ooi, Ji Jinn Foo Physics of Fluids, 2025 Efficient mixing at the microscales is essential for optimizing mass transfer and reaction rates in various microfluidic applications, underscoring the significance of comprehending and manipulating surface roughness to improve mixing performance. Surface morphology in microchannels is inherently influenced by fabrication and post-treatment. This study investigates the effects of three-dimensional (3D) Gaussian-generated random roughness on species homogenization. Nine roughness profiles, varying in (a) relative roughness (ε = 0.4%, 0.7%, 1.0%) and (b) correlation length (k = 10%, 20%, 30%), form the channel base at ReDh = 100. Using the lattice Boltzmann method, we examine the mixing efficiency (MI), velocity statistics, and spatial frequency. Higher ε enhances near-wall mixing, with a 5.7% MI increase for ε = 1.0% compared to 0.4%. Conversely, shorter correlation lengths create more rugged surfaces, increasing interfacial area for diffusion and thereby elevating the near-wall MI by 8.4% when k decreases from 30% to 10%. Spatial frequency analysis confirms that higher spatial frequencies (shorter spatial wavelengths, lower k) improve near-surface mixing. However, smoother surfaces (higher k) reduce global flow resistance, enhance central advective effects, and improve overall outlet mixing. Thus, for practical applications emphasizing outlet performance, lower ε and higher k yield superior results. This study not only advances our understanding of surface roughness parameters for fluid mixing in rough-walled microchannels and highlights the significance of spatial frequency characteristics but also offers valuable insights into optimizing mixing processes in diverse applications.
Augmentation of piezoelectric thin-film flapping velocimetry turbulence strength detection via machine learning Ted Sian Lee, Ean Hin Ooi, Wei Sea Chang, Ji Jinn Foo Physics of Fluids, 2025 Qualitatively evaluating the fundamental mechanical characteristics of square-fractal-grid (SFG)-generated turbulent flow using piezoelectric thin-film flapping velocimetry (PTFV) is rather time-consuming. More importantly, its sensitivity in detecting high-frequency, fine-scale turbulent fluctuations is constrained by high-speed camera specifications. To reduce dependency on high-speed imaging in future PTFV implementations, regression models are trained with supervised machine learning to determine the correlation between piezoelectric-generated voltage V and the corresponding local equivalent flow velocity fluctuation. Using V and thin-film tip deflection δ data as predictors and responses, respectively, Trilayered Neural Network (TNN) emerges as the best-performing model compared to linear regression, regression trees, support vector machines, Gaussian process regression, and ensembles of trees. TNN models trained on data from the (i) lower quarter, (ii) bottom left corner, and (iii) central opening of the SFG-grid provide accurate predictions of insert-induced centerline streamwise and cross-sectional equivalent lateral turbulence intensity and root mean square-δ, with average errors not exceeding 5%. The output predicted from the V response, which considers small-scale turbulence fluctuations across the entire thin-film surface, better expresses the equivalent lateral integral length scale (38% smaller) and turbulence forcing (270% greater), particularly at the bottom left corner of SFG where small-scale eddies are significant. Furthermore, the TNN model effectively captures the occasional extensive excitation forces from large-scale turbulent eddies, resulting in a more balanced force distribution. In short, this study paves the path for comprehensive and expedited flow dynamics characterization and turbulence forcing detection via PTFV, with potential deployment in high Reynolds number flows generated by various grid configurations.
Optimisation of open pit slope design considering groundwater effects using particle swarm optimisation and scaled boundary finite element method Dakshith Ruvin Wijesinghe, Ethmadalage Perera, Ean Hin Ooi, Sundararajan Natarajan, Taghi Sherizadeh, Ean Tat Ooi Engineering Analysis with Boundary Elements, 2024 Slopes are a crucial structures in open pit mines. Their design has implications on the economic, safety and environmental operation of the mining industry. Designing stable slopes can be challenging due to the complexities introduced by the stratigraphy and hydrology of the strata. With rising commodity costs and inflation rates, mining operating costs are increasing. Reducing operational costs is necessary for mining industries to remain competitive. While steepening the pit slope can decrease stripping materials and save money, it also increases the risk associated with slope surges. Therefore, optimising slopes is crucial for both financial and safety reasons. Numerical models such as the finite element method experience challenges in mesh generation of heterogeneous systems characterised by varying material properties and stratigraphies. Moreover, the need for repetitive geometry update necessitates recursive mesh regeneration that increases the computational burden. Moreover, previous slope optimisation studies focus solely on dry conditions. To consider the complex condition of hydrology along with heterogeneity in the soil stratigraphy, this study develops an optimisation procedure by combining the particle swarm optimisation algorithm and the scaled boundary finite element with an image-based meshing technique to optimise slopes with groundwater and achieve the desired factor of safety (FoS). The method changes the slope design parameters and the phreatic surface of groundwater simultaneously, considering user-defined parameters while iteratively re-meshing the optimisation processes. Several cases are presented, demonstrating the optimisation of bench width, bench angle, backfill parameters, and groundwater pumping levels.
Anisotropic Heat Transfer in a Fibrous Membrane with Hierarchically Assembled 2D Materials Yu Du, Fangzheng Zhen, Siyuan Ding, Yueni Zhong, Peixuan Li, Ke Zhan, Miheng Dong, Zhijun Guo, Weiren Fan, Ooi Ean Hin, Baofu Ding, Ruiping Zou, Ling Qiu, Aibing Yu, Minsu Liu ACS Applied Materials and Interfaces, 2024 Effective heat redistribution in specific directions is vital for advanced thermal management, significantly enhancing device performance by optimizing spatial heat configurations. We have designed and fabricated a hierarchical fibrous membrane that enables precise heat directing. By integrating hierarchical structure design with the anisotropic thermal conductivity of two-dimensional (2D) materials, we developed a fibrous membrane for anisotropic heat transfer. Such a structure is fabricated by aligning a 1D structured fiber in the 2D plane to achieve anisotropy at each scale level. The fiber units, where 2D nanosheets circumferentially and axially aligned, achieved a high axial thermal conductivity of 16.8 W·m-1·K-1 and advanced heat directing ability, confirmed by characterizations and simulations. The assembled membrane demonstrated an exceptional tensile strength (365 MPa) and high thermal conductivity (10.5 W·m-1·K-1) along the fiber axis. Our membranes are seen as a refined model for thermal management materials, combining the benefits of heat spreaders and thermal interface materials, thus being proficient in directing heat along programmed pathways. A practical wireless charging cooling demonstration illustrated this. Our methodology also proved versatile with different 2D fillers and various geometries. This research presents a method to achieve precise heat directing at the material's level, facilitating the systematic design of thermal management in electronics.
Enhancing sonothrombolysis outcomes with dual-frequency ultrasound: Insights from an in silico microbubble dynamics study Zhi Qi Tan, Ean H Ooi, Yeong Shiong Chiew, Ji Jinn Foo, Yin Kwee Ng, Ean Tat Ooi Computers in Biology and Medicine, 2024 Sonothrombolysis is a technique that employs the ultrasound waves to break down the clot. Recent studies have demonstrated significant improvement in the treatment efficacy when combining two ultrasound waves of different frequencies. Nevertheless, the findings remain conflicted on the ideal frequency pairing that leads to an optimal treatment outcome. Existing experimental studies are constrained by the limited range of frequencies that can be investigated, while numerical studies are typically confined to spherical microbubble dynamics, thereby restricting the scope of the analysis. To overcome this, the present study investigated the microbubble dynamics caused by the different combinations of ultrasound frequencies. This was carried out using computational modelling as it enables the visualisation of the microbubble behaviour, which is difficult in experimental studies due to the opacity of blood. The results showed that the pairings of two ultrasound waves with low frequencies generally produced stronger cavitation and higher flow-induced shear stress on the clot surface. However, one should avoid the frequency pairings that are integer multipliers of each other, i.e., frequency ratio of 1/3, 1/2 and 2, as they led to resultant wave with low pressure amplitude that weakened the cavitation. At 0.5 + 0.85 MHz, the microbubble caused the highest shear stress of 60.5 kPa, due to its large translational distance towards the clot. Although the pressure threshold for inertial cavitation was reduced using dual-frequency ultrasound, the impact of the high-speed jet can only be realised when the microbubble travelled close to the clot. The results obtained from the present study provide groundwork for deeper understanding on the microbubble dynamics during dual-frequency sonothrombolysis, which is of paramount importance for its optimisations and the subsequent clinical translation.
Gold nanorod–assisted theranostic solution for nonvisible residual disease in bladder cancer Paolo Armanetti, Irene Locatelli, Chiara Venegoni, Elisa Alchera, Beatrice Campanella, Filippo Pederzoli, Mirko Maturi, Erica Locatelli, Silvia Tortorella, Flavio Curnis, Angelo Corti, Roberta Lucianò, Massimo Onor, Andrea Salonia, Francesco Montorsi, Marco Moschini, Viktor Popov, Jithin Jose, Mauro Comes Franchini, Ean Hin Ooi, Luca Menichetti, Massimo Alfano Proceedings of the National Academy of Sciences of the United States of America, 2024
A numerical study on the no-touch bipolar radiofrequency ablation∗ Shelley Yap, Jason K. K. Cheong, Ean H Ooi, Iman Y Liao, Ji J. Foo, Shalini R. Nair, Ahmad F. Mohd Ali Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society EMBS, 2019
