IMITATE: Clinical Prior Guided Hierarchical Vision-Language Pre-training.

In the field of medical Vision-Language Pretraining (VLP), significant efforts have been devoted to deriving text and image features from both clinical reports and associated medical images. However, most existing methods may have overlooked the opportunity in leveraging the inherent hierarchical structure of clinical reports, which are generally split into 'findings' for descriptive content and 'impressions' for conclusive observation. Instead of utilizing this rich, structured format, current medical VLP approaches often simplify the report into either a unified entity or fragmented tokens. 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. Comprehensive experimental results highlight the advantages of integrating the hierarchical structure of medical reports for vision-language alignment.

Liquid-Liquid Dispersion Performance Prediction and Uncertainty Quantification Using Recurrent Neural Networks.

We 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.

Spectral Cross-Domain Neural Network With Soft-Adaptive Threshold Spectral Enhancement.

Electrocardiography (ECG) signals can be considered as multivariable time series (TS). The state-of-the-art ECG data classification approaches, based on either feature engineering or deep learning techniques, treat separately spectral and time domains in machine learning systems. No spectral-time domain communication mechanism inside the classifier model can be found in current approaches, leading to difficulties in identifying complex ECG forms. In this article, we proposed a novel deep learning model named spectral cross-domain neural network (SCDNN) with a new block called soft-adaptive threshold spectral enhancement (SATSE), to simultaneously reveal the key information embedded in spectral and time domains inside the neural network. More precisely, the domain-cross information is captured by a general convolutional neural network (CNN) backbone, and different information sources are merged by a self-adaptive mechanism to mine the connection between time and spectral domains. In SATSE, the knowledge from time and spectral domains is extracted via the fast Fourier transformation (FFT) with soft trainable thresholds in modified sigmoid functions. The proposed SCDNN is tested with several classification tasks implemented on the public ECG databases PTB-XL and CPSC2018. SCDNN outperforms the state-of-the-art approaches with a low computational cost regarding a variety of metrics in all classification tasks on both databases, by finding appropriate domains from the infinite spectral mapping. The convergence of the trainable thresholds in the spectral domain is also numerically investigated in this article. The robust performance of SCDNN provides a new perspective to exploit knowledge across deep learning models from time and spectral domains. The code repository can be found: https://github.com/DL-WG/SCDNN-TS.

Real-time updating of dynamic social networks for COVID-19 vaccination strategies.

Vaccination 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.

Accurate identification and measurement of the precipitate area by two-stage deep neural networks in novel chromium-based alloys.

The performance of advanced materials for extreme environments is underpinned by their microstructure, such as the size and distribution of nano- to micro-sized reinforcing phase(s). Chromium-based superalloys are a recently proposed alternative to conventional face-centred-cubic superalloys for high-temperature applications, e.g., Concentrated Solar Power. Their development requires the determination of precipitate volume fraction and size distribution using Electron Microscopy (EM), as these properties are crucial for the thermal stability and mechanical properties of chromium superalloys. Traditional approaches to EM image processing utilise filtering with a fixed contrast threshold, leads to weak robustness to background noise and poor generalisability to different materials. It also requires an enormous amount of time for manual object measurements on large datasets. Efficient and accurate object detection and segmentation are therefore highly desired to accelerate the development of novel materials like chromium-based superalloys. To address these bottlenecks, based on YOLOv5 and SegFormer structures, this study proposes an end-to-end, two-stage deep learning scheme, DT-SegNet, to perform object detection and segmentation for EM images. The proposed approach can thus benefit from the training efficiency of CNNs at the detection stage (i.e., a small number of training images required) and the accuracy of the ViT at the segmentation stage. Extensive numerical experiments demonstrate that the proposed DT-SegNet significantly outperforms the state-of-the-art segmentation tools offered by Weka and ilastik regarding a large number of metrics, including accuracy, precision, recall and F1-score. This model forms a useful tool to aid alloy development microstructure examinations, and offers significant advantages to address the large datasets associated with high-throughput alloy development approaches.

Analyzing drop coalescence in microfluidic devices with a deep learning generative model.

Predicting drop coalescence based on process parameters is crucial for experimental design in chemical engineering. However, predictive models can suffer from the lack of training data and more importantly, the label imbalance problem. In this study, we propose the use of deep learning generative models to tackle this bottleneck by training the predictive models using generated synthetic data. A novel generative model, named double space conditional variational autoencoder (DSCVAE) is developed for labelled tabular data. By introducing label constraints in both the latent and the original space, DSCVAE is capable of generating consistent and realistic samples compared to the standard conditional variational autoencoder (CVAE). Two predictive models, namely random forest and gradient boosting classifiers, are enhanced on synthetic data and their performances are evaluated based on real experimental data. Numerical results show that a considerable improvement in prediction accuracy can be achieved by using synthetic data and the proposed DSCVAE clearly outperforms the standard CVAE. This research clearly provides more insights into handling imbalanced data for classification problems, especially in chemical engineering.

Surfactant-laden droplet size prediction in a flow-focusing microchannel: a data-driven approach.

The control of droplet formation and size using microfluidic devices is a critical operation for both laboratory and industrial applications, e.g. in micro-dosage. Surfactants can be added to improve the stability and control the size of the droplets by modifying their interfacial properties. In this study, a large-scale data set of droplet size was obtained from high-speed imaging experiments conducted on a flow-focusing microchannel where aqueous surfactant-laden droplets were generated in silicone oil. Three types of surfactants were used including anionic, cationic and non-ionic at concentrations below and above the critical micelle concentration (CMC). To predict the final droplet size as a function of flow rates, surfactant type and concentration of surfactant, two data-driven models were built. Using a Bayesian regularised artificial neural network and XGBoost, these models were initially based on four inputs (flow rates of the two phases, interfacial tension at equilibrium and the normalised surfactant concentration). The mean absolute percentage errors (MAPE) show that data-driven models are more accurate (MAPE = 3.9%) compared to semi-empirical models (MAPE = 11.4%). To overcome experimental difficulties in acquiring accurate interfacial tension values under some conditions, both models were also trained with reduced inputs by removing the interfacial tension. The results show again a very good prediction of the droplet diameter. Finally, over 10 000 synthetic data were generated, based on the initial data set, with a Variational Autoencoder (VAE). The high-fidelity of the extended synthetic data set highlights that this method can be a quick and low-cost alternative to study microdroplet formation in future lab on a chip applications, where experimental data may not be readily available.

Ensemble latent assimilation with deep learning surrogate model: application to drop interaction in a microfluidics device.

A major challenge in the field of microfluidics is to predict and control drop interactions. This work develops an image-based data-driven model to forecast drop dynamics based on experiments performed on a microfluidics device. Reduced-order modelling techniques are applied to compress the recorded images into low-dimensional spaces and alleviate the computational cost. Recurrent neural networks are then employed to build a surrogate model of drop interactions by learning the dynamics of compressed variables in the reduced-order space. The surrogate model is integrated with real-time observations using data assimilation. In this paper we developed an ensemble-based latent assimilation algorithm scheme which shows an improvement in terms of accuracy with respect to the previous approaches. This work demonstrates the possibility to create a reliable data-driven model enabling a high fidelity prediction of drop interactions in microfluidics device. The performance of the developed system is evaluated against experimental data (i.e., recorded videos), which are excluded from the training of the surrogate model. The developed scheme is general and can be applied to other dynamical systems.

Digital twins based on bidirectional LSTM and GAN for modelling the COVID-19 pandemic.

The outbreak of the coronavirus disease 2019 (COVID-19) has now spread throughout the globe infecting over 150 million people and causing the death of over 3.2 million people. Thus, there is an urgent need to study the dynamics of epidemiological models to gain a better understanding of how such diseases spread. While epidemiological models can be computationally expensive, recent advances in machine learning techniques have given rise to neural networks with the ability to learn and predict complex dynamics at reduced computational costs. Here we introduce two digital twins of a SEIRS model applied to an idealised town. The SEIRS model has been modified to take account of spatial variation and, where possible, the model parameters are based on official virus spreading data from the UK. We compare predictions from one digital twin based on a data-corrected Bidirectional Long Short-Term Memory network with predictions from another digital twin based on a predictive Generative Adversarial Network. The predictions given by these two frameworks are accurate when compared to the original SEIRS model data. Additionally, these frameworks are data-agnostic and could be applied to towns, idealised or real, in the UK or in other countries. Also, more compartments could be included in the SEIRS model, in order to study more realistic epidemiological behaviour.

An overview of methods of fine and ultrafine particle collection for physicochemical characterisation and toxicity assessments.

Particulate matter (PM) is a crucial health risk factor for respiratory and cardiovascular diseases. The smaller size fractions, ≤2.5 µm (PM2.5; fine particles) and ≤0.1 µm (PM0.1; ultrafine particles), show the highest bioactivity but acquiring sufficient mass for in vitro and in vivo toxicological studies is challenging. We review the suitability of available instrumentation to collect the PM mass required for these assessments. Five different microenvironments representing the diverse exposure conditions in urban environments are considered in order to establish the typical PM concentrations present. The highest concentrations of PM2.5 and PM0.1 were found near traffic (i.e. roadsides and traffic intersections), followed by indoor environments, parks and behind roadside vegetation. We identify key factors to consider when selecting sampling instrumentation. These include PM concentration on-site (low concentrations increase sampling time), nature of sampling sites (e.g. indoors; noise and space will be an issue), equipment handling and power supply. Physicochemical characterisation requires micro- to milli-gram quantities of PM and it may increase according to the processing methods (e.g. digestion or sonication). Toxicological assessments of PM involve numerous mechanisms (e.g. inflammatory processes and oxidative stress) requiring significant amounts of PM to obtain accurate results. Optimising air sampling techniques are therefore important for the appropriate collection medium/filter which have innate physical properties and the potential to interact with samples. An evaluation of methods and instrumentation used for airborne virus collection concludes that samplers operating cyclone sampling techniques (using centrifugal forces) are effective in collecting airborne viruses. We highlight that predictive modelling can help to identify pollution hotspots in an urban environment for the efficient collection of PM mass. This review provides guidance to prepare and plan efficient sampling campaigns to collect sufficient PM mass for various purposes in a reasonable timeframe.

An epidemiological modelling approach for COVID-19 via data assimilation.

The global pandemic of the 2019-nCov requires the evaluation of policy interventions to mitigate future social and economic costs of quarantine measures worldwide. We propose an epidemiological model for forecasting and policy evaluation which incorporates new data in real-time through variational data assimilation. We analyze and discuss infection rates in the UK, US and Italy. We furthermore develop a custom compartmental SIR model fit to variables related to the available data of the pandemic, named SITR model, which allows for more granular inference on infection numbers. We compare and discuss model results which conducts updates as new observations become available. A hybrid data assimilation approach is applied to make results robust to initial conditions and measurement errors in the data. We use the model to conduct inference on infection numbers as well as parameters such as the disease transmissibility rate or the rate of recovery. The parameterisation of the model is parsimonious and extendable, allowing for the incorporation of additional data and parameters of interest. This allows for scalability and the extension of the model to other locations or the adaption of novel data sources.