Search

The persistent challenge of medical image synthesis posed by the scarcity of annotated data and the need to synthesize `missing modalities' for multi-modal analysis, underscored the imperative development of effective synthesis methods. Recently, the combination of Low-Rank Adaptation (LoRA) with latent diffusion models (LDMs) has emerged as a viable approach for efficiently adapting pre-trained large language models, in the medical field. However, the direct application of LoRA assumes uniform ranking across all linear layers, overlooking the significance of different weight matrices, and leading to sub-optimal outcomes. Prior works on LoRA prioritize the reduction of trainable parameters, and there exists an opportunity to further tailor this adaptation process to the intricate demands of medical image synthesis. In response, we present SeLoRA, a Self-Expanding Low-Rank Adaptation Module, that dynamically expands its ranking across layers during training, strategically placing additional ranks on crucial layers, to allow the model to elevate synthesis quality where it matters most. The proposed method not only enables LDMs to fine-tune on medical data efficiently but also empowers the model to achieve improved image quality with minimal ranking. The code of our SeLoRA method is publicly available on https://anonymous.4open.science/r/SeLoRA-980D ., Comment: Project Page: https://yuchen20.github.io/SeLoRA.github.io/

Federated Learning (FL) has emerged as a promising solution to address the limitations of centralised machine learning (ML) in oncology, particularly in overcoming privacy concerns and harnessing the power of diverse, multi-center data. This systematic review synthesises current knowledge on the state-of-the-art FL in oncology, focusing on breast, lung, and prostate cancer. Distinct from previous surveys, our comprehensive review critically evaluates the real-world implementation and impact of FL on cancer care, demonstrating its effectiveness in enhancing ML generalisability, performance and data privacy in clinical settings and data. We evaluated state-of-the-art advances in FL, demonstrating its growing adoption amid tightening data privacy regulations. FL outperformed centralised ML in 15 out of the 25 studies reviewed, spanning diverse ML models and clinical applications, and facilitating integration of multi-modal information for precision medicine. Despite the current challenges identified in reproducibility, standardisation and methodology across studies, the demonstrable benefits of FL in harnessing real-world data and addressing clinical needs highlight its significant potential for advancing cancer research. We propose that future research should focus on addressing these limitations and investigating further advanced FL methods, to fully harness data diversity and realise the transformative power of cutting-edge FL in cancer care., Comment: 5 Figures, 3 Tables, 1 Supplementary Table

Immunofluorescent (IF) imaging is crucial for visualizing biomarker expressions, cell morphology and assessing the effects of drug treatments on sub-cellular components. IF imaging needs extra staining process and often requiring cell fixation, therefore it may also introduce artefects and alter endogenouous cell morphology. Some IF stains are expensive or not readily available hence hindering experiments. Recent diffusion models, which synthesise high-fidelity IF images from easy-to-acquire brightfield (BF) images, offer a promising solution but are hindered by training instability and slow inference times due to the noise diffusion process. This paper presents a novel method for the conditional synthesis of IF images directly from BF images along with cell segmentation masks. Our approach employs a Residual Diffusion process that enhances stability and significantly reduces inference time. We performed a critical evaluation against other image-to-image synthesis models, including UNets, GANs, and advanced diffusion models. Our model demonstrates significant improvements in image quality (p<0.05 in MSE, PSNR, and SSIM), inference speed (26 times faster than competing diffusion models), and accurate segmentation results for both nuclei and cell bodies (0.77 and 0.63 mean IOU for nuclei and cell true positives, respectively). This paper is a substantial advancement in the field, providing robust and efficient tools for cell image analysis.

Cell imaging assays utilizing fluorescence stains are essential for observing sub-cellular organelles and their responses to perturbations. Immunofluorescent staining process is routinely in labs, however the recent innovations in generative AI is challenging the idea of IF staining are required. This is especially true when the availability and cost of specific fluorescence dyes is a problem to some labs. Furthermore, staining process takes time and leads to inter-intra technician and hinders downstream image and data analysis, and the reusability of image data for other projects. Recent studies showed the use of generated synthetic immunofluorescence (IF) images from brightfield (BF) images using generative AI algorithms in the literature. Therefore, in this study, we benchmark and compare five models from three types of IF generation backbones, CNN, GAN, and diffusion models, using a publicly available dataset. This paper not only serves as a comparative study to determine the best-performing model but also proposes a comprehensive analysis pipeline for evaluating the efficacy of generators in IF image synthesis. We highlighted the potential of deep learning-based generators for IF image synthesis, while also discussed potential issues and future research directions. Although generative AI shows promise in simplifying cell phenotyping using only BF images with IF staining, further research and validations are needed to address the key challenges of model generalisability, batch effects, feature relevance and computational costs.

Generative AI has revolutionised visual content editing, empowering users to effortlessly modify images and videos. However, not all edits are equal. To perform realistic edits in domains such as natural image or medical imaging, modifications must respect causal relationships inherent to the data generation process. Such image editing falls into the counterfactual image generation regime. Evaluating counterfactual image generation is substantially complex: not only it lacks observable ground truths, but also requires adherence to causal constraints. Although several counterfactual image generation methods and evaluation metrics exist, a comprehensive comparison within a unified setting is lacking. We present a comparison framework to thoroughly benchmark counterfactual image generation methods. We integrate all models that have been used for the task at hand and expand them to novel datasets and causal graphs, demonstrating the superiority of Hierarchical VAEs across most datasets and metrics. Our framework is implemented in a user-friendly Python package that can be extended to incorporate additional SCMs, causal methods, generative models, and datasets for the community to build on.

Deep learning (DL) has substantially enhanced natural language processing (NLP) in healthcare research. However, the increasing complexity of DL-based NLP necessitates transparent model interpretability, or at least explainability, for reliable decision-making. This work presents a thorough scoping review of explainable and interpretable DL in healthcare NLP. The term "eXplainable and Interpretable Artificial Intelligence" (XIAI) is introduced to distinguish XAI from IAI. Different models are further categorized based on their functionality (model-, input-, output-based) and scope (local, global). Our analysis shows that attention mechanisms are the most prevalent emerging IAI technique. The use of IAI is growing, distinguishing it from XAI. The major challenges identified are that most XIAI does not explore "global" modelling processes, the lack of best practices, and the lack of systematic evaluation and benchmarks. One important opportunity is to use attention mechanisms to enhance multi-modal XIAI for personalized medicine. Additionally, combining DL with causal logic holds promise. Our discussion encourages the integration of XIAI in Large Language Models (LLMs) and domain-specific smaller models. In conclusion, XIAI adoption in healthcare requires dedicated in-house expertise. Collaboration with domain experts, end-users, and policymakers can lead to ready-to-use XIAI methods across NLP and medical tasks. While challenges exist, XIAI techniques offer a valuable foundation for interpretable NLP algorithms in healthcare., Comment: This paper has been accepted by Computational and Structural Biotechnology Journal

Clinical decision making from magnetic resonance imaging (MRI) combines complementary information from multiple MRI sequences (defined as 'modalities'). MRI image registration aims to geometrically 'pair' diagnoses from different modalities, time points and slices. Both intra- and inter-modality MRI registration are essential components in clinical MRI settings. Further, an MRI image processing pipeline that can address both afine and non-rigid registration is critical, as both types of deformations may be occuring in real MRI data scenarios. Unlike image classification, explainability is not commonly addressed in image registration deep learning (DL) methods, as it is challenging to interpet model-data behaviours against transformation fields. To properly address this, we incorporate Grad-CAM-based explainability frameworks in each major component of our unsupervised multi-modal and multi-organ image registration DL methodology. We previously demonstrated that we were able to reach superior performance (against the current standard Syn method). In this work, we show that our DL model becomes fully explainable, setting the framework to generalise our approach on further medical imaging data.

Medical imaging is a key component in clinical diagnosis, treatment planning and clinical trial design, accounting for almost 90% of all healthcare data. CNNs achieved performance gains in medical image analysis (MIA) over the last years. CNNs can efficiently model local pixel interactions and be trained on small-scale MI data. The main disadvantage of typical CNN models is that they ignore global pixel relationships within images, which limits their generalisation ability to understand out-of-distribution data with different 'global' information. The recent progress of Artificial Intelligence gave rise to Transformers, which can learn global relationships from data. However, full Transformer models need to be trained on large-scale data and involve tremendous computational complexity. Attention and Transformer compartments (Transf/Attention) which can well maintain properties for modelling global relationships, have been proposed as lighter alternatives of full Transformers. Recently, there is an increasing trend to co-pollinate complementary local-global properties from CNN and Transf/Attention architectures, which led to a new era of hybrid models. The past years have witnessed substantial growth in hybrid CNN-Transf/Attention models across diverse MIA problems. In this systematic review, we survey existing hybrid CNN-Transf/Attention models, review and unravel key architectural designs, analyse breakthroughs, and evaluate current and future opportunities as well as challenges. We also introduced a comprehensive analysis framework on generalisation opportunities of scientific and clinical impact, based on which new data-driven domain generalisation and adaptation methods can be stimulated.

Synthetic images generated from deep generative models have the potential to address data scarcity and data privacy issues. The selection of synthesis models is mostly based on image quality measurements, and most researchers favor synthetic images that produce realistic images, i.e., images with good fidelity scores, such as low Fr\'echet Inception Distance (FID) and high Peak Signal-To-Noise Ratio (PSNR). However, the quality of synthetic images is not limited to fidelity, and a wide spectrum of metrics should be evaluated to comprehensively measure the quality of synthetic images. In addition, quality metrics are not truthful predictors of the utility of synthetic images, and the relations between these evaluation metrics are not yet clear. In this work, we have established a comprehensive set of evaluators for synthetic images, including fidelity, variety, privacy, and utility. By analyzing more than 100k chest X-ray images and their synthetic copies, we have demonstrated that there is an inevitable trade-off between synthetic image fidelity, variety, and privacy. In addition, we have empirically demonstrated that the utility score does not require images with both high fidelity and high variety. For intra- and cross-task data augmentation, mode-collapsed images and low-fidelity images can still demonstrate high utility. Finally, our experiments have also showed that it is possible to produce images with both high utility and privacy, which can provide a strong rationale for the use of deep generative models in privacy-preserving applications. Our study can shore up comprehensive guidance for the evaluation of synthetic images and elicit further developments for utility-aware deep generative models in medical image synthesis., Comment: 10 pages, 4 figures, MICCAI Early Acceptance

As a pragmatic data augmentation tool, data synthesis has generally returned dividends in performance for deep learning based medical image analysis. However, generating corresponding segmentation masks for synthetic medical images is laborious and subjective. To obtain paired synthetic medical images and segmentations, conditional generative models that use segmentation masks as synthesis conditions were proposed. However, these segmentation mask-conditioned generative models still relied on large, varied, and labeled training datasets, and they could only provide limited constraints on human anatomical structures, leading to unrealistic image features. Moreover, the invariant pixel-level conditions could reduce the variety of synthetic lesions and thus reduce the efficacy of data augmentation. To address these issues, in this work, we propose a novel strategy for medical image synthesis, namely Unsupervised Mask (UM)-guided synthesis, to obtain both synthetic images and segmentations using limited manual segmentation labels. We first develop a superpixel based algorithm to generate unsupervised structural guidance and then design a conditional generative model to synthesize images and annotations simultaneously from those unsupervised masks in a semi-supervised multi-task setting. In addition, we devise a multi-scale multi-task Fr\'echet Inception Distance (MM-FID) and multi-scale multi-task standard deviation (MM-STD) to harness both fidelity and variety evaluations of synthetic CT images. With multiple analyses on different scales, we could produce stable image quality measurements with high reproducibility. Compared with the segmentation mask guided synthesis, our UM-guided synthesis provided high-quality synthetic images with significantly higher fidelity, variety, and utility ($p<0.05$ by Wilcoxon Signed Ranked test)., Comment: 12 pages, 11 figures, accepted by IEEE TMI

Current network inference algorithms fail to generate graphs with edges that can explain whole sequences of node interactions in a given dataset or trace. To quantify how well an inferred graph can explain a trace, we introduce feasibility, a novel quality criterion, and suggest that it is linked to the result's accuracy. In addition, we propose CEM-*, a network inference method that guarantees 100% feasibility given online social media traces, which is a non-trivial extension of the Expectation-Maximization algorithm developed by Newman (2018). We propose a set of linear optimization updates that incorporate a set of auxiliary variables and a set of feasibility constraints; the latter takes into consideration all the hidden paths that are possible between users based on their timestamps of interaction and guide the inference toward feasibility. We provide two CEM-* variations, that assume either an Erdos Renyi (ER) or a Stochastic Block Model (SBM) prior for the underlying graph's unknown distribution. Extensive experiments on one synthetic and one real-world Twitter dataset show that for both priors CEM-* can generate a posterior distribution of graphs that explains the whole trace while being closer to the ground truth. As an additional benefit, the use of the SBM prior infers and clusters users simultaneously during optimization. CEM-* outperforms baseline and state-of-the-art methods in terms of feasibility, run-time, and precision of the inferred graph and communities. Finally, we propose a heuristic to adapt the inference to lower feasibility requirements and show how it can affect the precision of the result.

Modern security proofs of quantum key distribution (QKD) must take finite-size effects and composable aspects into consideration. This is also the case for continuous-variable (CV) protocols which are based on the transmission and detection of bosonic coherent states. In this paper, we refine and advance the previous theory in this area providing a more rigorous formulation for the composable key rate of a generic CV-QKD protocol. Thanks to these theoretical refinements, our general formulas allow us to prove more optimistic key rates with respect to previous literature., Comment: Results largely extended. Improves theory from arXiv:2203.00706, arXiv:2010.04168, and arXiv:1912.11418

The extensive emergence of big data techniques has led to an increasing interest in the development of change-point detection algorithms that can perform well in a multivariate, possibly high-dimensional setting. In the current paper, we propose a new method for the consistent estimation of the number and location of multiple generalized change-points in multivariate, possibly high-dimensional, noisy data sequences. The number of change-points is allowed to increase with the sample size and the dimensionality of the given data sequence. Having a number of univariate signals, which constitute the unknown multivariate signal, our algorithm can deal with general structural changes; we focus on changes in the mean vector of a multivariate piecewise-constant signal, as well as changes in the linear trend of any of the univariate component signals. Our proposed algorithm, labeled Multivariate Isolate-Detect (MID), allows for consistent change-point detection in the presence of frequent changes of possibly small magnitudes in a computationally fast way., Comment: 38 pages, 6 figures. arXiv admin note: text overlap with arXiv:1901.10852

Circuit depth reduction is of critical importance for quantum chemistry simulations on current and near term quantum computers. This issue is tackled by introducing a chemically aware strategy for the Unitary Coupled Cluster ansatz. The objective is to use the chemical description of a system to aid in the synthesis of a quantum circuit. We combine this approach with two flavours of Symmetry Verification for the reduction of experimental noise. These method enable the use of System Model H1 for a 6-qubit QSE (Quantum Subspace Expansion). We present (i) calculations to obtain CH4 optical spectra; (ii) an atmospheric gas reaction simulation involving $[$CH$^{\cdot}_3$--H--OH$^{\cdot}]^{\ddagger}$. Using our chemically aware UCC state-preparation strategy in tandem with state of the art symmetry verification methods, we improve device yield for CH4 at 6-qubits. This is demonstrated by a 90% improvement in two-qubit gate count and reduction in relative error to 0.2% for electronic energy calculated on System Model H1.

Online social platforms have become central in the political debate. In this context, the existence of echo chambers is a problem of primary relevance. These clusters of like-minded individuals tend to reinforce prior beliefs, elicit animosity towards others and aggravate the spread of misinformation. We study this phenomenon on a Twitter dataset related to the 2017 French presidential elections and propose a method to tackle it with content recommendations. We use a quadratic program to find optimal recommendations that maximise the diversity of content users are exposed to, while still accounting for their preferences. Our method relies on a theoretical model that can sufficiently describe how content flows through the platform. We show that the model provides good approximations of empirical measures and demonstrate the effectiveness of the optimisation algorithm at mitigating the echo chamber effect on this dataset, even with limited budget for recommendations.

The performance of a practical continuous-variable (CV) quantum key distribution (QKD) protocol depends significantly, apart from the loss and noise of the quantum channel, on the post-processing steps which lead to the extraction of the final secret key. A critical step is the reconciliation process, especially when one assumes finite-size effects in a composable framework. Here, we focus on the Gaussian-modulated coherent-state protocol with heterodyne detection in a high signal-to-noise ratio regime. We simulate the quantum communication process and we post-process the output data by applying parameter estimation, error correction (using high-rate, non-binary low-density parity-check codes), and privacy amplification. This allows us to study the performance for practical implementations of the protocol and optimize the parameters connected to the steps above. We also present an associated Python library performing the steps above., Comment: 13 pages, 5 figures, 2 tables

Gaussian networks are fundamental objects in network information theory. Here many senders and receivers are connected by physically motivated Gaussian channels while auxiliary Gaussian components, such as Gaussian relays, are entailed. Whilst the theoretical backbone of classical Gaussian networks is well established, the quantum analogue is yet immature. Here, we theoretically tackle composable security of arbitrary Gaussian quantum networks (quantum networks), with generally untrusted nodes, in the finite-size regime. We put forward a general methodology for parameter estimation, which is only based on the data shared by the remote end-users. Taking a chain of identical quantum links as an example, we further demonstrate our study. Additionally, we find that the key rate of a quantum amplifier-assisted chain can ideally beat the fundamental repeaterless limit with practical block sizes. However, this objective is practically questioned leading the way to new network/chain designs., Comment: 12 pages, 5 figures

Magnetic Resonance Imaging (MRI) typically recruits multiple sequences (defined here as "modalities"). As each modality is designed to offer different anatomical and functional clinical information, there are evident disparities in the imaging content across modalities. Inter- and intra-modality affine and non-rigid image registration is an essential medical image analysis process in clinical imaging, as for example before imaging biomarkers need to be derived and clinically evaluated across different MRI modalities, time phases and slices. Although commonly needed in real clinical scenarios, affine and non-rigid image registration is not extensively investigated using a single unsupervised model architecture. In our work, we present an un-supervised deep learning registration methodology which can accurately model affine and non-rigid trans-formations, simultaneously. Moreover, inverse-consistency is a fundamental inter-modality registration property that is not considered in deep learning registration algorithms. To address inverse-consistency, our methodology performs bi-directional cross-modality image synthesis to learn modality-invariant latent rep-resentations, while involves two factorised transformation networks and an inverse-consistency loss to learn topology-preserving anatomical transformations. Overall, our model (named "FIRE") shows improved performances against the reference standard baseline method on multi-modality brain 2D and 3D MRI and intra-modality cardiac 4D MRI data experiments., Comment: 38 pages, 7 figures, will be published in Sensors journal by MDPI

In this paper the problem of a plane strain hydraulic fracture in elasto-plastic material is analyzed. The analysis is based on Finite Element Method computations of crack propagation and closure to simulate the Mini-Frac calibration test. The basic trends of fracture evolution are identified for both, the propagation and the closure stage in elasto-plastic materials and compared with the results obtained for purely elastic deformation of rock. Pressure decline analysis is conducted to identify the closure stress and leak-off coefficient. The results show that the standard techniques of pressure decline analysis can be adopted also for the case where the fractured material deforms inelastically.

Network inference is the process of deciding what is the true unknown graph underlying a set of interactions between nodes. There is a vast literature on the subject, but most known methods have an important drawback: the inferred graph is not guaranteed to explain every interaction from the input trace. We consider this an important issue since such inferred graph cannot be used as input for applications that require a reliable estimate of the true graph. On the other hand, a graph having trace feasibility guarantees can help us better understand the true (hidden) interactions that may have taken place between nodes of interest. The inference of such graph is the goal of this paper. Firstly, given an activity log from a social network, we introduce a set of constraints that take into consideration all the hidden paths that are possible between the nodes of the trace, given their timestamps of interaction. Then, we develop a nontrivial modification of the Expectation-Maximization algorithm by Newman [1], that we call Constrained-EM, which incorporates the constraints and a set of auxiliary variables into the inference process to guide it towards the feasibility of the trace. Experimental results on real-world data from Twitter confirm that Constrained-EM generates a posterior distribution of graphs that explains all the events observed in the trace while presenting the desired properties of a scale-free, small-world graph. Our method also outperforms established methods in terms of feasibility and quality of the inferred graph., Comment: Proceedings of the 2021 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), Nov 2021, The Hague, Netherlands

In this paper a problem of numerical simulation of hydraulic fractures is considered. An efficient algorithm of solution is proposed for the plain strain model of hydraulic fracturing. The algorithm utilizes a FEM based subroutine to compute deformation of the fractured material. The flow of generalized Newtonian fluid in the fracture is modelled in the framework of lubrication theory. In this way, the architecture of the computational scheme is relatively simple and enables one to deal with advanced cases of the fractured material properties and configurations as well as various rheological models of fluid. In particular, the problems of poroelasticity, plasticity and spatially varying properties of the fractured material can be analyzed. The accuracy and efficiency of the proposed algorithm are verified against analytical benchmark solutions. The algorithm capabilities are demonstrated using the example of the hydraulic fracture propagating in complex geological settings.

In this paper the problem of a plane strain hydraulic fracture propagating in an elasto-plastic material is analyzed. A new stress redistribution model for the proximity of the fracture tip is formulated and a resulting plasticity-dependent crack propagation condition is introduced. A modified variant of the KGD problem that accounts for the plastic deformations in the near-tip zone only is proposed. It is demonstrated that this model can be a credible substitute for the full elasto-plastic hydraulic fracture problem in the case of moderate plastic deformation. The crack-tip shielding effect introduced by the plastic deformation is quantified.

We investigate the performance of Gaussianmodulated coherent-state QKD protocols in the presence of canonical attacks, which are collective Gaussian attacks resulting in Gaussian channels described by one of the possible canonical forms. We present asymptotic key rates and then we extend the results to the finite-size regime using a recently-developed toolbox for composable security., Comment: 6 pages, 4 figures

Continuous-variable (CV) quantum key distribution (QKD) employs the quadratures of a bosonic mode to establish a secret key between two remote parties, and this is usually achieved via a Gaussian modulation of coherent states. The resulting secret key rate depends not only on the loss and noise in the communication channel, but also on a series of data processing steps that are needed for transforming shared correlations into a final string of secret bits. Here we consider a Gaussian-modulated coherent-state protocol with homodyne detection in the general setting of composable finite-size security. After simulating the process of quantum communication, the output classical data is post-processed via procedures of parameter estimation, error correction, and privacy amplification. In particular, we analyze the high signal-to-noise regime which requires the use of high-rate (non-binary) low-density parity check codes. We implement all these steps in a Python-based library that allows one to investigate and optimize the protocol parameters to be used in practical experimental implementations of short-range CV-QKD., Comment: Accepted version. The code for data-processing is available on github at https://github.com/softquanta/homCVQKD