Synthetic Data Research Articles

Utilization of machine learning methods for prediction of acute and late rectal toxicity due to curative prostate radiotherapy.

Rectal toxicity is one of the primary dose-limiting side effects of prostate cancer radiotherapy, and consequential impairment on quality of life in these patients with long survival is an important problem. In this study, we aimed to evaluate the possibility of predicting rectal toxicity with artificial intelligence model which was including certain dosimetric parameters. One hundred and thirty-seven patients with a diagnosis of prostate cancer who received curative radiotherapy for prostate +/- pelvic lymphatics were included in the study. The association of the clinical data and dosimetric data between early and late rectal toxicity reported during follow-up was evaluated. The sample size was increased to 274 patients by synthetic data generation method. To determine suitable models, 15 models were studied with machine learning algorithms using Python 2.3, Pycaret library. Random forest classifier was used with to detect active variables. The area under the curve and accuracy were found to be 0.89-0.97 and 95%-99%, respectively, with machine learning algorithms. The sensitivity values for acute and toxicity were found to be 0.95 and 0.99, respectively. Early or late rectal toxicity can be predicted with a high probability via dosimetric and physical data and machine learning algorithms of patients who underwent prostate +/- pelvic radiotherapy. The fact that rectal toxicity can be predicted before treatment, which may result in limiting the dose and duration of treatment, makes us think that artificial intelligence can enter our daily practice in a short time in this sense.

Optimizing compressive strength prediction using adversarial learning and hybrid regularization

The infrastructure industry consumes natural resources and produces construction waste, which has a detrimental impact on the environment. To mitigate these adverse effects and reduce raw material consumption, waste materials can be repurposed to achieve sustainability. However, recycled materials deteriorate the intrinsic properties of concrete. A suitable ratio of natural resources and recycled aggregates can produce the desired compressive strength. Compiling sufficient data in civil engineering laboratories to make reliable conclusions is time-consuming and costly. Therefore, this research proposes a novel approach for predicting compressive strengths using limited data. The generative adversarial network was employed to generate synthetic data. Hybrid training, utilizing either conventional loss or heuristic loss, prevents the model from overfitting by adaptively adjusting the regularization term. Random noise from a multivariate normal distribution is embedded heuristically into the training samples to capture intricate data variations. Sensitivity analysis indicated that the size of recycled coarse aggregate and water are the most significant features, aligning with their correlations. Interestingly, superplasticizer, density of recycled coarse aggregate, and water absorption ratio of recycled coarse aggregate contributed significantly to predictions despite their low correlations. The propounded method outperforms random forest, support vector regression, artificial neural network, and adaptive boosting by scoring a mean squared error of 7.97, a root mean squared error of 2.82, a mean absolute error of 2.13, and a coefficient of determination of 0.96. These results suggest that the proposed technique can effectively contribute to sustainable construction practices by accurately predicting compressive strengths.

Deep joint inversion of multiple geophysical data with U-net reparameterization

Joint inversion integrates various geophysical data in a unified framework to reveal different subsurface properties. Existing methods often require predefined correlations through mathematical equations or training datasets, heavily relying on prior information. Additionally, integrating multi-physics data with different sensitivities is challenging. To address these limitations, we propose a novel approach that utilizes U-net to automatically establish correlations during inversion, which yields consistent inverted models across different resolutions. Different properties are represented by U-net parameters and solved by minimizing multi-modality data misfits simultaneously. We validate our approach through structure-constrained electromagnetic inversion, joint seismic and gravity inversion, and joint electromagnetic, seismic, and gravity inversion with synthetic data. This study affirms the feasibility of leveraging machine intelligence to automatically integrate multiple geophysical data for precise subsurface characterization.

An Attention-Based Deep Learning Model for Phase-Resolved Wave Prediction

Abstract Phase-resolved wave prediction capability, even if only over two wave periods in advance, is of value for optimal control of wave energy converters, resulting in a dramatic increase in power generation efficiency. Previous studies on wave-by-wave predictions have shown that an artificial neural network (ANN) model can outperform the traditional linear wave theory-based model in terms of both prediction accuracy and prediction horizon when using synthetic wave data. However, the prediction performance of ANN models is significantly reduced by the varying wave conditions and buoy positions that occur in the field. To overcome these limitations, a novel wave prediction method is developed based on the neural network with an attention mechanism. This study validates the new model using wave data measured at sea. The model utilizes past time histories of three Sofar Spotter wave buoys at upwave locations to predict the vertical motion of a Datawell Waverider-4 at a downwave location. The results show that the attention-based neural network model is capable of capturing the slow variation in the displacement of the buoys, which reduces the prediction error compared to a standard ANN and long short-term memory model.

Leveraging Machine Learning for Advanced Nanoscale X-ray Analysis: Unmixing Multicomponent Signals and Enhancing Chemical Quantification.

Energy dispersive X-ray (EDX) spectroscopy in the transmission electron microscope is a key tool for nanomaterials analysis, providing a direct link between spatial and chemical information. However, using it for precisely determining chemical compositions presents challenges of noisy data from low X-ray yields and mixed signals from phases that overlap along the electron beam trajectory. Here, we introduce a novel method, non-negative matrix factorization based pan-sharpening (PSNMF), to address these limitations. Leveraging the Poisson nature of EDX spectral noise and binning operations, PSNMF retrieves high-quality phase spectral and spatial signatures via consecutive factorizations. After validating PSNMF with synthetic data sets of different noise levels, we illustrate its effectiveness on two distinct experimental cases: a nanomineralogical lamella, and supported catalytic nanoparticles. Not only does PSNMF obtain accurate phase signatures, but data sets reconstructed from the outputs have demonstrably lower noise and better fidelity than from the benchmark denoising method of principle component analysis.

A Pseudo-Spectral Method for Wall Shear Stress Estimation from Doppler Ultrasound Imaging in Coronary Arteries.

The Wall Shear Stress (WSS) is the component tangential to the boundary of the normal stress tensor in an incompressible fluid, and it has been recognized as a quantity of primary importance in predicting possible adverse events in cardiovascular diseases, in general, and in coronary diseases, in particular. The quantification of the WSS in patient-specific settings can be achieved by performing a Computational Fluid Dynamics (CFD) analysis based on patient geometry, or it can be retrieved by a numerical approximation based on blood flow velocity data, e.g., ultrasound (US) Doppler measurements. This paper presents a novel method for WSS quantification from 2D vector Doppler measurements. Images were obtained through unfocused plane waves and transverse oscillation to acquire both in-plane velocity components. These velocity components were processed using pseudo-spectral differentiation techniques based on Fourier approximations of the derivatives to compute the WSS. Our Pseudo-Spectral Method (PSM) is tested in two vessel phantoms, straight and stenotic, where a steady flow of 15mL/min is applied. The method is successfully validated against CFD simulations and compared against current techniques based on the assumption of a parabolic velocity profile. The PSM accurately detected Wall Shear Stress (WSS) variations in geometries differing from straight cylinders, and is less sensitive to measurement noise. In particular, when using synthetic data (noise free, e.g., generated by CFD) on cylindrical geometries, the Poiseuille-based methods and PSM have comparable accuracy; on the contrary, when using the data retrieved from US measures, the average error of the WSS obtained with the PSM turned out to be 3 to 9 times smaller than that obtained by state-of-the-art methods. The pseudo-spectral approach allows controlling the approximation errors in the presence of noisy data. This gives a more accurate alternative to the present standard and a less computationally expensive choice compared to CFD, which also requires high-quality data to reconstruct the vessel geometry.

MLAWSMOTE: Oversampling in Imbalanced Multi-label Classification with Missing Labels by Learning Label Correlation Matrix

Missing labels in multi-label datasets are a common problem, especially for minority classes, which are more likely to occur. This limitation hinders the performance of classifiers in identifying and extracting information from minority classes. Oversampling is an effective method for addressing imbalanced multi-label problems by generating synthetic instances to create a class-balanced dataset. However, the existing oversampling algorithms mainly focus on the location of the generated data, and there is a lack of design on how to complete the labels of the synthetic data. To address this issue, we propose MLAWSMOTE, a synthetic data generation algorithm based on matrix factorization weights. We introduce a weak supervised learning method in the oversampling method, optimize the weights of features and labels by using label correlation, and iteratively learn the ideal label weights. The mapping relationship between features and labels is learned from the dataset and the label correlation matrix. The oversampling ratio is defined based on the discrepancy between observed labels and the ideal label of synthetic instances. It mitigates the impact of missing minority labels on the model’s predictions. The labeling of synthetic instances is performed based on label prediction, and the potential labeling distribution is complemented. Experimental results on multiple multi-label datasets under different label missing ratios demonstrate the effectiveness of the proposed method in terms of ACC, Hamming loss, MacroF1 and MicroF1. In the validation of the four classifiers, MacroF1 decreased by 24.78%, 17.81%, 3.8% and 19.56%, respectively, with the increase of label loss rate. After applying MLAWSMOTE only decreased by 15.79%, 13.63%, 3.78% and 15.21%.

Synthetic Medical Imaging Generation with Generative Adversarial Networks for Plain Radiographs

In medical imaging, access to data is commonly limited due to patient privacy restrictions, and it can be difficult to acquire enough data in the case of rare diseases. The purpose of this investigation was to develop a reusable open-source synthetic image-generation pipeline, the GAN Image Synthesis Tool (GIST), that is easy to use as well as easy to deploy. The pipeline helps to improve and standardize AI algorithms in the digital health space by generating high quality synthetic image data that is not linked to specific patients. Its image generation capabilities include the ability to generate imaging of pathologies or injuries with low incidence rates. This improvement of digital health AI algorithms could improve diagnostic accuracy, aid in patient care, decrease medicolegal claims, and ultimately decrease the overall cost of healthcare. The pipeline builds on existing Generative Adversarial Networks (GANs) algorithms, and preprocessing and evaluation steps were included for completeness. For this work, we focused on ensuring the pipeline supports radiography, with a focus on synthetic knee and elbow X-ray images. In designing the pipeline, we evaluated the performance of current GAN architectures, studying the performance on available X-ray data. We show that the pipeline is capable of generating high-quality and clinically relevant images based on a lay person’s evaluation and the Fréchet Inception Distance (FID) metric.

An optimal transport approach to estimating causal effects via nonlinear difference-in-differences

Abstract We propose a nonlinear difference-in-differences (DiD) method to estimate multivariate counterfactual distributions in classical treatment and control study designs with observational data. Our approach sheds a new light on existing approaches like the changes-in-changes estimator and the classical semiparametric DiD estimator, and it also generalizes them to settings with multivariate heterogeneity in the outcomes. The main benefit of this extension is that it allows for arbitrary dependence between the coordinates of vector potential outcomes and includes higher-dimensional unobservables, something that existing methods cannot provide in general. We demonstrate its utility on both synthetic and real data. In particular, we revisit the classical Card & Krueger dataset, which reports fast food restaurant employment before and after a minimum wage increase. A reanalysis with our methodology suggests that these restaurants substitute full-time labor with part-time labor on aggregate in response to a minimum wage increase. This treatment effect requires estimation of the multivariate counterfactual distribution, an object beyond the scope of classical causal estimators previously applied to this data.

Multichannel seismic attenuation compensation and interpolation with curvelet sparse constraint of frequency-wavenumber spectrum

Abstract High-resolution exploration is hampered by seismic attenuation, caused by the viscosity and heterogeneity of underground media. Conventional single-channel attenuation compensation methods can increase the corresponding vertical resolution partially. While the lateral continuity of compensated seismic data is always ignored and the noise resistance can be improved. Therefore, multichannel attenuation compensation methods were proposed, including the algorithm with sparse curvelet coefficient constraint of the time-space (t-x) data. However, nonstationary seismic data may be affected by irregular missing traces in field cases, which severely degrades its lateral continuity and negatively affects the performance of multichannel attenuation compensation. Therefore, we concentrate on simultaneous multichannel attenuation compensation and missing trace interpolation in a unified framework. Based on the inversion framework of sparsity promotion, we propose an approach for simultaneous multichannel compensation and interpolation, utilizing sparse curvelet coefficient constraint of the recovered principal frequency-wavenumber (f-k) spectrum. The size of the principal f-k spectrum is reduced by at least half compared with that of the corresponding t-x band-limited data. It significantly reduces the computational expense of curvelet transform-based processing. Synthetic and field data experiments validate the effectiveness of the proposed method in efficiency improvement with consistent performance when compared to the multichannel method with sparse curvelet coefficient constraint of the t-x data in improving the vertical resolution and lateral continuity. Furthermore, we have discussed a potential acceleration strategy based on random sampling in the normal compensation issue.

Anomaly Detection for Atomic Clocks using Unsupervised Machine Learning Algorithms

Abstract Atomic clocks are the most accurate timekeeping instruments in the world and play a crucial role in numerous applications, notably in satellite navigation. Nonetheless, the accuracy of these clock signals is subject to anomalies, highlighting the importance of anomaly characterization and detection. In our study, we identify three types of anomalies in the phase data of atomic clocks—outliers, phase jumps and inflection points, and also frequency jumps in the frequency data, which corresponds to the inflection points. The novelty of our work lies in the application of machine learning methods for atomic clock anomaly detection, as opposed to traditional algorithms. We introduce a variety of unsupervised anomaly detection(AD) and change point detection (CPD) algorithms from the realms of statistics and machine learning, propose a customized thresholding function for AD, and evaluate AD and CPD algorithms on both synthetic and real-world atomic clock data. Our experiments reveal that the Subsequence Local Outlier Factor (Sub-LOF) algorithm achieves excellent performance in detecting outliers and phase jumps, but struggles with inflection points. Crucially, we found that the application of a CPD algorithm specifically for detecting the frequency jumps in frequency data substantially improves performance. Our study adopts unsupervised machine learning methods for anomaly detection provides robust detection strategies, thereby contributing an expanded understanding of the use of machine learning for clock data monitoring.

Coincident learning for unsupervised anomaly detection of scientific instruments

Abstract Anomaly detection is an important task for complex scientific experiments and other complex systems (e.g. industrial facilities, manufacturing), where failures in a sub-system can lead to lost data, poor performance, or even damage to components. While scientific facilities generate a wealth of data, labeled anomalies may be rare (or even nonexistent), and expensive to acquire. Unsupervised approaches are therefore common and typically search for anomalies either by distance or density of examples in the input feature space (or some associated low-dimensional representation). This paper presents a novel approach called coincident learning for anomaly detection (CoAD), which is specifically designed for multi-modal tasks and identifies anomalies based on coincident behavior across two different slices of the feature space. We define an unsupervised metric, F ^ β , out of analogy to the supervised classification F β statistic. CoAD uses F ^ β to train an anomaly detection algorithm on unlabeled data, based on the expectation that anomalous behavior in one feature slice is coincident with anomalous behavior in the other. The method is illustrated using a synthetic outlier data set and a MNIST-based image data set, and is compared to prior state-of-the-art on two real-world tasks: a metal milling data set and our motivating task of identifying RF station anomalies in a particle accelerator.

Parameter Identification by Eigenfeature Analysis: Application to 2D Kuramoto-Sivashinsky Surface Models

Abstract We have developed a system that makes it possible to derive parameters of a Kuramoto-Sivashinsky (KS) model from a single given two-dimensional profile of surface structures, such as those produced by ion and plasma irradiation. The numerical method is inspired by well-known approaches to facial recognition. Starting from a scaled version of a KS Model to describe surface erosion, a training set of surface profiles is created. Each profile is assigned an appropriate feature in Fourier space and a Singular Value Decomposition is used to determine an orthogonal set of eigenfeatures that allow each profile to be assigned a point in the space of this basis and to determine the distances between them. It turns out that the profiles belonging to different model parameters are clearly separated from each other in this feature space, which enables very good identification. We explain the basic relationships using a synthetic data set and discuss the possibilities for applications to experimental results.

Mass Conservative Time-Series GAN for Synthetic Extreme Flood-Event Generation: Impact on Probabilistic Forecasting Models

The lack of data on flood events poses challenges in flood management. In this paper, we propose a novel approach to enhance flood-forecasting models by utilizing the capabilities of Generative Adversarial Networks (GANs) to generate synthetic flood events. We modified a time-series GAN by incorporating constraints related to mass conservation, energy balance, and hydraulic principles into the GAN model through appropriate regularization terms in the loss function and by using mass conservative LSTM in the generator and discriminator models. In this way, we can improve the realism and physical consistency of the generated extreme flood-event data. These constraints ensure that the synthetic flood-event data generated by the GAN adhere to fundamental hydrological principles and characteristics, enhancing the accuracy and reliability of flood-forecasting and risk-assessment applications. PCA and t-SNE are applied to provide valuable insights into the structure and distribution of the synthetic flood data, highlighting patterns, clusters, and relationships within the data. We aimed to use the generated synthetic data to supplement the original data and train probabilistic neural runoff model for forecasting multi-step ahead flood events. t-statistic was performed to compare the means of synthetic data generated by TimeGAN with the original data, and the results showed that the means of the two datasets were statistically significant at 95% level. The integration of time-series GAN-generated synthetic flood events with real data improved the robustness and accuracy of the autoencoder model, enabling more reliable predictions of extreme flood events. In the pilot study, the model trained on the augmented dataset with synthetic data from time-series GAN shows higher NSE and KGE scores of NSE = 0.838 and KGE = 0.908, compared to the NSE = 0.829 and KGE = 0.90 of the sixth hour ahead, indicating improved accuracy of 9.8% NSE in multistep-ahead predictions of extreme flood events compared to the model trained on the original data alone. The integration of synthetic training datasets in the probabilistic forecasting improves the model’s ability to achieve a reduced Prediction Interval Normalized Average Width (PINAW) for interval forecasting, yet this enhancement comes with a trade-off in the Prediction Interval Coverage Probability (PICP).

CryoSamba: self-supervised deep volumetric denoising for cryo-electron tomography data.

Cryogenic electron tomography (cryo-ET) has rapidly advanced as a high-resolution imaging tool for visualizing subcellular structures in 3D with molecular detail. Direct image inspection remains challenging due to inherent low signal-to-noise ratios (SNR). We introduce CryoSamba, a self-supervised deep learning-based model designed for denoising cryo-ET images. CryoSamba enhances single consecutive 2D planes in tomograms by averaging motion-compensated nearby planes through deep learning interpolation, effectively mimicking increased exposure. This approach amplifies coherent signals and reduces high-frequency noise, substantially improving tomogram contrast and SNR. CryoSamba operates on 3D volumes without needing pre-recorded images, synthetic data, labels or annotations, noise models, or paired volumes. CryoSamba suppresses high-frequency information less aggressively than do existing cryo-ET denoising methods, while retaining real information, as shown both by visual inspection and by Fourier shell correlation analysis of icosahedrally symmetric virus particles. Thus, CryoSamba enhances the analytical pipeline for direct 3D tomogram visual interpretation.

Full-waveform automatic location of small seismic events in an underground mine using synthetic strain Greens tensors

A reliable method to locate microseismic evets is an important objective for routine mine seismic monitoring. We developed an algorithm that can automatically and reliably locate an event from a single uniaxial trace. Signals created by microseismic events are recorded as seismograms that consist of direct P- and S-waves as well as the waves reflected off the underground excavations. We take advantage of the extra complexity these reflections introduce as they make the signal more unique. In our investigation we cross-correlated unaltered synthetic seismograms against trial waveforms created from a linear combination of appropriate waveforms from a library of numerically calculated strain Green's tensors. We use a single uniaxial synthetic seismogram to invert for the source location, six moment tensor components, and frequency. The resulting cost functions are constructed as maximal correlations at each node in the mine grid through the true source x-planes. This function is maximized by differential evolution to simultaneously yield the source parameters. Sensitivity tests such as shifting the mine plans and altering the background velocities of the media are investigated to test the robustness of the method. We contaminate the data with 40% white noise to test the resolution against simulated real recorded seismograms. The method is demonstrated using synthetic data calculated with a realistic underground 3D velocity model.

LUIE: Learnable physical model-guided underwater image enhancement with bi-directional unsupervised domain adaptation

Recently, learning-based underwater enhancement (UIE) methods have made considerable progress, significantly benefiting downstream tasks such as underwater semantic segmentation and underwater depth estimation. Most existing unsupervised UIE methods utilize the atmospheric image formation model to decompose underwater images into background color, transmission map, and scene radiance. However, they rely on simplified physical models for estimating the transmission map, over-simplifying its complex formation, which results in imprecise modeling of underwater scattering effects. Additionally, supervised UIE methods heavily depend on synthetic data or ground truth, leading to limited generalization capabilities due to the substantial domain gap presented in different underwater scenarios. To tackle these challenges, we propose a Learnable physical model-guided unsupervised domain adaptation framework for Underwater Image Enhancement, dubbed LUIE. LUIE learns to predict background light, depth, and scene radiance from an underwater image. We incorporate a learnable network to estimate the transmission map based on the predicted depth map. To minimize the inter-domain gap between synthetic and real underwater images, we introduce a bi-directional domain adaptation method that alternates the background light from each domain. Experimental results demonstrate the effectiveness of our proposed method compared to existing approaches, and high-level experiment results validate that our enhanced underwater results. Experiments in real-world settings on underwater ROVs platform with NVIDIA Jetson AGX Xavier further confirm the effectiveness and efficiency of our work.

Bayesian reconstruction of 3D particle positions in high-seeding density flows

Abstract Measuring particles' three-dimensional (3D) positions using multi-camera images in fluid dynamics is critical for resolving spatiotemporally complex flows like turbulence and mixing. However, current methods are prone to errors due to camera noise, optical configuration and experimental setup limitations, and high seeding density, which compound to create fake measurements (ghost particles) and add noise and error to velocity estimations. We introduce a Bayesian Volumetric Reconstruction (BVR) method, addressing these challenges by using probability theory to estimate uncertainties in particle position predictions. Our method assumes a uniform distribution of particles within the reconstruction volume and employs a model mapping particle positions to observed camera images. We utilize variational inference with a modified loss function to determine the posterior distribution over particle positions. Key features include a penalty term to reduce ghost particles, provision of uncertainty bounds, and scalability through subsampling. In tests with synthetic data and four cameras, BVR achieved 95% accuracy with less than 3% ghost particles and an RMS error under 0.3 pixels at a density of 0.1 particles per pixel (ppp). This shows 57% to 97% less ghost particles and 1.3 to 2 times lower RMS error than standard MART and IPR reconstructions. In an experimental Poiseuille flow measurement, our method closely matched the theoretical solution. Additionally, in a complex cerebral aneurysm basilar tip geometry flow experiment, our reconstructions were dense and consistent with observed flow patterns. Our BVR method effectively reconstructs particle positions in complex 3D flows, particularly in situations with high particle image densities and camera distortions. It distinguishes itself by providing quantifiable uncertainty estimates and scaling efficiently for larger image dimensions, making it applicable across a range of fluid flow scenarios.

Bridge the gap between simulated and real-world data in optical fiber mode decomposition for accuracy improvement: A deep learning-based co-learning framework with visual similarity-based matching

In the area of optical fiber mode decomposition (MD), data-driven deep-learning (DL) algorithms have achieved remarkable strides with unlimited theoretically synthetic data, but the differences between theoretical and real-world data in the DL context are rarely investigated. The two differences, namely the information bottleneck and the domain discrepancy, between the simulation and realistic domain remain significant obstacles that limit the practical deployment of DL-based MD approaches in real-world optical systems. The paper aims to mitigate their influence on the accuracy limitation and the accuracy degradation of DL-based MD. First, we present a DL-based co-learning framework for MD for the first time. With the co-learning process, the knowledge of the teacher model learned from rich input information sources (near-field, far-field, and phase-distribution optical images) is transferred to the student model with the only near-field input source, which breaks the information bottleneck by allowing the student model to learn more knowledge in a simulation context. Furthermore, to settle the domain discrepancy between ideal data and practical observation, we propose a visual similarity-based matching to assign real values of modal coefficients to the practical optical images whose ground truth values cannot be easily acquired, which enables the transfer learning from the synthetic data to the real-world data for accuracy enhancement. Simulated and experimental results show that the proposed co-learning framework with visual similarity-based matching has realized high precision and robustness in MD simulated and practical tasks. By bridging the gap between theoretical and practical data, our study offers novel perspectives on DL-based MD practice.

From minimum-viable-products to full models: a step-wise development of diagnostic forward models in support of design, analysis and modelling on the ST40 tokamak

Like most magnetic confined fusion experiments, the ST40 tokamak started off with a small subset of diagnostics and gradually increased the diagnostic set to include more complex and comprehensive systems. To make the most of each operational phase, forward models of various diagnostics are used and developed to aid design, provide consistency-checks during commissioning, test analysis methods, and build workflows to constrain high-level parameters to inform interpretation, theory and modelling. For new models and new analysis workflows, minimum-viable-products are released early, and their complexity is increased in a step-wise manner, facilitating the support of all programme phases on multiple parallel applications, while enabling learning opportunities and feedback loops. In this contribution we review the philosophy, scope and architecture of the framework under development. We discuss the details of some forward models, with examples on how they are used to aid diagnostic design, to investigate analysis methodologies through synthetic data, and how they are embedded in experimental analysis workflows. We compare previously published experimental results with new, more advanced analysis workflows employing more recent, detailed models and new diagnostic data, providing confirmation of the published material from the 2021–22 experimental campaign.