Results
- Showing results for:
- Reset all filters
Search results
-
Journal articleShao P, Jiang Y, Kovalchuk NM, et al., 2026,
Explainability of AI predictive models for drop coalescence in microfluidics channels: Experimental validation
, CHEMICAL ENGINEERING SCIENCE, Vol: 320, ISSN: 0009-2509 -
Journal articleCheng S, Bocquet M, Ding W, et al., 2025,
Machine learning for modelling unstructured grid data in computational physics: A review
, INFORMATION FUSION, Vol: 123, ISSN: 1566-2535- Cite
- Citations: 3
-
Journal articleNathanael K, Cheng S, Kovalchuk NM, et al., 2025,
Optimisation of microfluidic synthesis of silver nanoparticles via data-driven inverse modelling
, CHEMICAL ENGINEERING RESEARCH & DESIGN, Vol: 216, Pages: 523-530, ISSN: 0263-8762- Cite
- Citations: 1
-
Journal articleFan H, Cheng S, de Nazelle AJ, et al., 2025,
ViTAE-SL: a vision transformer-based autoencoder and spatial interpolation learner for field reconstruction
, Computer Physics Communications, Vol: 308, ISSN: 0010-4655Reliable and accurate reconstruction for large-scale and complex physical fields in real-time from limited observations has been a longstanding challenge. In recent years, sensors have been increasingly deployed in numerous physical systems. However, the locations of these sensors can shift over time, such as with mobile sensors, or when sensors are deployed and removed. These sparse and randomly located sensors further exacerbate the difficulty of reconstructing the physical field. In this paper, we present a new deep learning model called Vision Transformer-based Autoencoder (ViTAE) for reconstructing large-scale and complex fields. The proposed network structure is based on a novel core design: vision transformer encoder and Convolutional Neural Network (CNN) decoder. First, we split a two-dimensional field into patches and developed a vision transformer encoder to transfer patches into latent representations. We then reshape the linear latent representations to patches before concatenation, along with a CNN decoder, to reconstruct the field. The proposed model is tested in four different numerical experiments, using generated synthetic data, spatially distributed PM2.5 data, Computational Fluid Dynamics (CFD) simulation data and National Oceanic and Atmospheric Administration (NOAA) sea surface temperature data. The numerical results highlight the strength of ViTAE-SL compared to Kriging and state-of-the-art deep-learning models with significantly higher reconstruction accuracy, computational efficiency, and robust scaling behavior.
-
Journal articleLiu C, Cheng S, Shi M, et al., 2025,
IMITATE: clinical prior guided hierarchical vision-language pre-training
, IEEE Transactions on Medical Imaging, Vol: 44, Pages: 519-529, ISSN: 0278-0062In medical Vision-Language Pre-training (VLP), significant work focuses on extracting text and image features from clinical reports and medical images. Yet, existing methods may overlooked the potential of the natural hierarchical structure in clinical reports, typically divided into ‘findings’ for description and ‘impressions’ for conclusions. Current VLP approaches tend to oversimplify these reports into a single entity or fragmented tokens, ignoring this structured format. In this work, we propose a novel clinical prior guided VLP framework named IMITATE to learn the structure information from medical reports with hierarchical vision-language alignment. The framework derives multi-level visual features from the chest X-ray (CXR) images and separately aligns these features with the descriptive and the conclusive text encoded in the hierarchical medical report. Furthermore, a new clinical-informed contrastive loss is introduced for cross-modal learning, which accounts for clinical prior knowledge in formulating sample correlations in contrastive learning. The proposed model, IMITATE, outperforms baseline VLP methods across six different datasets, spanning five medical imaging downstream tasks. Experimental results show benefits of using hierarchical structures in medical reports for VLP. Code: https://github.com/cheliu-computation/IMITATE-TMI2024.
-
Journal articleWang K, Bertoli G, Cheng S, et al., 2025,
AI-Empowered Latent Four-dimensional Variational Data Assimilation for River Discharge Forecasting
, IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, Vol: 18, Pages: 24676-24689, ISSN: 1939-1404 -
Journal articlePlatt R, AlZayer Z, Arcucci R, et al., 2025,
Teaching Ourselves to See: A Direct Method for Denoising CRISM Hyperspectral Data
, IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, Vol: 63, ISSN: 0196-2892 -
Journal articleBasha N, Arcucci R, Angeli P, et al., 2024,
Machine learning and physics-driven modelling and simulation of multiphase systems
, International Journal of Multiphase Flow, Vol: 179, ISSN: 0301-9322We highlight the work of a multi-university collaborative programme, PREMIERE (PREdictive Modelling with QuantIfication of UncERtainty for MultiphasE Systems), which is at the intersection of multi-physics and machine learning, aiming to enhance predictive capabilities in complex multiphase flow systems across diverse length and time scales. Our contributions encompass a variety of approaches, including the Design of Experiments for nanoparticle synthesis optimisation, Generalised Latent Assimilation models for drop coalescence prediction, Bayesian regularised artificial neural networks, eXtreme Gradient Boosting for microdroplet formation prediction, and a sub-sampling based adversarial neural network for predicting slug flow behaviour in two-phase pipe flows. Additionally, we introduce a generalised latent assimilation technique, Long Short-Term Memory networks for sequence forecasting mixing performance in stirred and static mixers, active learning via Bayesian optimisation to recover coalescence model parameters for high current density electrolysers, Gaussian process regression for drop size distribution predictions for sprays, and acoustic emission signal inversion using gradient boosting machines to characterise particle size distribution in fluidised beds. We also offer perspectives on the development of a shape optimisation framework that leverages the use of a multi-fidelity multiphase emulator. The results presented have applications in chemical synthesis, microfluidics, product manufacturing, and green hydrogen generation.
-
Journal articleLiang F, Valdes JP, Cheng S, et al., 2024,
Liquid-liquid dispersion performance prediction and uncertainty quantification using recurrent neural networks
, Industrial and Engineering Chemistry Research, Vol: 63, Pages: 7853-7875, ISSN: 0888-5885We demonstrate the application of a recurrent neural network (RNN) to perform multistep and multivariate time-series performance predictions for stirred and static mixers as exemplars of complex multiphase systems. We employ two network architectures in this study, fitted with either long short-term memory and gated recurrent unit cells, which are trained on high-fidelity, three-dimensional, computational fluid dynamics simulations of the mixer performance, in the presence and absence of surfactants, in terms of drop size distributions and interfacial areas as a function of system parameters; these include physicochemical properties, mixer geometry, and operating conditions. Our results demonstrate that while it is possible to train RNNs with a single fully connected layer more efficiently than with an encoder-decoder structure, the latter is shown to be more capable of learning long-term dynamics underlying dispersion metrics. Details of the methodology are presented, which include data preprocessing, RNN model exploration, and methods for model performance visualization; an ensemble-based procedure is also introduced to provide a measure of the model uncertainty. The workflow is designed to be generic and can be deployed to make predictions in other industrial applications with similar time-series data.
-
Journal articleHu J, Zhu K, Cheng S, et al., 2024,
Explainable AI models for predicting drop coalescence in microfluidics device
, CHEMICAL ENGINEERING JOURNAL, Vol: 481, ISSN: 1385-8947 -
Journal articleKalaiarasan G, Kumar P, Tomson M, et al., 2024,
Particle number size distribution in three different microenvironments of London
, Atmosphere, Vol: 15, ISSN: 2073-4433We estimated the particle number distributions (PNDs), particle number concentrations (PNCs), physicochemical characteristics, meteorological effects, and respiratory deposition doses (RDD) in the human respiratory tract for three different particle modes: nucleation (N6–30), accumulation (N30–300), and coarse (N300–10,000) modes. This study was conducted in three different microenvironments (MEs) in London (indoor, IN; traffic intersection, TI; park, PK) measuring particles in the range of 6 nm–10,000 nm using an electrical low-pressure impactor (ELPI+). Mean PNCs were 1.68 ± 1.03 × 104 #cm−3, 7.00 ± 18.96 × 104 #cm−3, and 0.76 ± 0.95 × 104 #cm−3 at IN, TI, and PK, respectively. The PNDs were high for nucleation-mode particles at the TI site, especially during peak traffic hours. Wind speeds ranging from 0 to 6 ms−1 exhibit higher PNCs for nucleation- and accumulation-mode particles at TI and PK sites. Physicochemical characterisation shows trace metals, including Fe, O, and inorganic elements, that were embedded in a matrix of organic material in some samples. Alveolar RDD was higher for the nucleation and accumulation modes than the coarse-mode particles. The chemical signatures from the physicochemical characterisation indicate the varied sources at different MEs. These findings enhance our understanding of the different particle profiles at each ME and should help devise ways of reducing personal exposure at each ME.
-
Conference paperChen Y, Liu C, Liu X, et al., 2024,
BIMCV-R: A Landmark Dataset for 3D CT Text-Image Retrieval
, 27th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 124-134, ISSN: 0302-9743 -
Journal articleXia Z, Ma K, Cheng S, et al., 2023,
Accurate identification and measurement of the precipitate area by two-stage deep neural networks in novel chromium-based alloys
, PHYSICAL CHEMISTRY CHEMICAL PHYSICS, Vol: 25, Pages: 15970-15987, ISSN: 1463-9076- Author Web Link
- Cite
- Citations: 1
-
Journal articleCheng S, Quilodran-Casas C, Ouala S, et al., 2023,
Machine Learning With Data Assimilation and Uncertainty Quantification for Dynamical Systems: A Review
, IEEE-CAA JOURNAL OF AUTOMATICA SINICA, Vol: 10, Pages: 1361-1387, ISSN: 2329-9266- Author Web Link
- Cite
- Citations: 4
-
Journal articleZhong C, Cheng S, Kasoar M, et al., 2023,
Reduced-order digital twin and latent data assimilation for global wildfire prediction
, NATURAL HAZARDS AND EARTH SYSTEM SCIENCES, Vol: 23, Pages: 1755-1768, ISSN: 1561-8633- Author Web Link
- Cite
- Citations: 1
-
Journal articleNathanael K, Cheng S, Kovalchuk NM, et al., 2023,
Optimization of microfluidic synthesis of silver nanoparticles: A generic approach using machine learning
, CHEMICAL ENGINEERING RESEARCH & DESIGN, Vol: 193, Pages: 65-74, ISSN: 0263-8762- Author Web Link
- Cite
- Citations: 3
-
Journal articleArcucci R, Xiao D, Fang F, et al., 2023,
A reduced order with data assimilation model: Theory and practice
, Computers and Fluids, Vol: 257, Pages: 1-12, ISSN: 0045-7930Numerical simulations are extensively used as a predictive tool to better understand complex air flows and pollution transport on the scale of individual buildings, city blocks and entire cities. Fast-running Non-Intrusive Reduced Order Model (NIROM) for predicting the turbulent air flows has been proved to be an efficient method to provide numerical forecasting results. However, due to the reduced space on which the model operates, the solution includes uncertainties. Additionally, any computational methodology contributes to uncertainty due to finite precision and the consequent accumulation and amplification of round-off errors. Taking into account these uncertainties is essential for the acceptance of any numerical simulation. In this paper we combine the NIROM method with Data Assimilation (DA), the main question is how to incorporate data (e.g. from physical measurements) in models in a suitable way, in order to improve model predictions and quantify prediction uncertainty. Here, the focus is on the prediction of nonlinear dynamical systems (the classical application example being weather forecasting). DA is an uncertainty quantification technique used to incorporate observed data into a prediction model in order to improve numerical forecasted results. The Reduced Order Data Assimilation (RODA) model we propose in this paper achieves both efficiency and accuracy including Variational DA into NIROM. The model we present is applied to the pollutant dispersion within an urban environment.
-
Journal articleCheng S, Pain CC, Guo Y-K, et al., 2023,
Real-time updating of dynamic social networks for COVID-19 vaccination strategies.
, J Ambient Intell Humaniz Comput, Pages: 1-14, ISSN: 1868-5137Vaccination strategy is crucial in fighting the COVID-19 pandemic. Since the supply is still limited in many countries, contact network-based interventions can be most powerful to set an efficient strategy by identifying high-risk individuals or communities. However, due to the high dimension, only partial and noisy network information can be available in practice, especially for dynamic systems where contact networks are highly time-variant. Furthermore, the numerous mutations of SARS-CoV-2 have a significant impact on the infectious probability, requiring real-time network updating algorithms. In this study, we propose a sequential network updating approach based on data assimilation techniques to combine different sources of temporal information. We then prioritise the individuals with high-degree or high-centrality, obtained from assimilated networks, for vaccination. The assimilation-based approach is compared with the standard method (based on partially observed networks) and a random selection strategy in terms of vaccination effectiveness in a SIR model. The numerical comparison is first carried out using real-world face-to-face dynamic networks collected in a high school, followed by sequential multi-layer networks generated relying on the Barabasi-Albert model emulating large-scale social networks with several communities.
-
Journal articleChen S, Tai Z, Liu J, 2023,
Barriers, Facilitators, and Sustainers in Tai Ji Quan Practice: A Mixed-Methods RE-AIM Assessment of College Students Versus the General Population
, JOURNAL OF PHYSICAL ACTIVITY & HEALTH, Vol: 20, Pages: 239-249, ISSN: 1543-3080- Cite
- Citations: 11
-
Journal articleChen J, Anastasiou C, Cheng S, et al., 2023,
Computational fluid dynamics simulations of phase separation in dispersed oil-water pipe flows
, Chemical Engineering Science, Vol: 267, Pages: 1-18, ISSN: 0009-2509The separation of liquid–liquid dispersions in horizontal pipes is common in many industrial sectors. It remains challenging, however, to predict the separation characteristics of the flow evolution due to the complex flow mechanisms. In this work, Computational Fluid Dynamics (CFD) simulations of the silicone oil and water two-phase flow in a horizontal pipe are performed. Several cases are explored with different mixture velocities and oil fractions (15%-60%). OpenFOAM (version 8.0) is used to perform Eulerian-Eulerian simulations coupled with population balance models. The ‘blending factor’ in the multiphaseEulerFoam solver captures the retardation of the droplet rising and coalescing due to the complex flow behaviour in the dense packed layer (DPL). The blending treatment provides a feasible compensation mechanism for the mesoscale uncertainties of droplet flow and coalescence through the DPL and its adjacent layers. In addition, the influence of the turbulent dispersion force is also investigated, which can improve the prediction of the radial distribution of concentrations but worsen the separation characteristics along the flow direction. Although the simulated concentration distribution and layer heights agree with the experiments only qualitatively, this work demonstrates how improvements in drag and coalescence modelling can be made to enhance the prediction accuracy.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.