The Image Biomarker Standardisation Initiative (IBSI) aims to improve reproducibility of radiomics studies by standardising the computational process of extracting image biomarkers (features) from images. We have previously established reference values for 169 commonly used features, created a standard radiomics image processing scheme, and developed reporting guidelines for radiomic studies. However, several aspects are not standardised. Here we present a complete version of a reference manual on the use of convolutional filters in radiomics and quantitative image analysis. Filters, such as wavelets or Laplacian of Gaussian filters, play an important part in emphasising specific image characteristics such as edges and blobs. Features derived from filter response maps were found to be poorly reproducible. This reference manual provides definitions for convolutional filters, parameters that should be reported, reference feature values, and tests to verify software compliance with the reference standard.

Domain adaptation methods aim to bridge the gap between datasets by enabling knowledge transfer across domains, reducing the need for additional expert annotations. However, many approaches struggle with reliability in the target domain, an issue particularly critical in medical image segmentation, where accuracy and anatomical validity are essential. This challenge is further exacerbated in spatio-temporal data, where the lack of temporal consistency can significantly degrade segmentation quality, and particularly in echocardiography, where the presence of artifacts and noise can further hinder segmentation performance. To address these issues, we present RL4Seg3D, an unsupervised domain adaptation framework for 2D + time echocardiography segmentation. RL4Seg3D integrates novel reward functions and a fusion scheme to enhance key landmark precision in its segmentations while processing full-sized input videos. By leveraging reinforcement learning for image segmentation, our approach improves accuracy, anatomical validity, and temporal consistency while also providing, as a beneficial side effect, a robust uncertainty estimator, which can be used at test time to further enhance segmentation performance. We demonstrate the effectiveness of our framework on over 30,000 echocardiographic videos, showing that it outperforms standard domain adaptation techniques without the need for any labels on the target domain. Code is available at this https URL.

Accurate uncertainty estimation is a critical need for the medical imaging community. A variety of methods have been proposed, all direct extensions of classification uncertainty estimations techniques. The independent pixel-wise uncertainty estimates, often based on the probabilistic interpretation of neural networks, do not take into account anatomical prior knowledge and consequently provide sub-optimal results to many segmentation tasks. For this reason, we propose CRISP a ContRastive Image Segmentation for uncertainty Prediction method. At its core, CRISP implements a contrastive method to learn a joint latent space which encodes a distribution of valid segmentations and their corresponding images. We use this joint latent space to compare predictions to thousands of latent vectors and provide anatomically consistent uncertainty maps. Comprehensive studies performed on four medical image databases involving different modalities and organs underlines the superiority of our method compared to state-of-the-art approaches.

Generative Adversarial Networks (GANs) have become increasingly powerful, generating mind-blowing photorealistic images that mimic the content of datasets they were trained to replicate. One recurrent theme in medical imaging is whether GANs can also be effective at generating workable medical data as they are for generating realistic RGB images. In this paper, we perform a multi-GAN and multi-application study to gauge the benefits of GANs in medical imaging. We tested various GAN architectures from basic DCGAN to more sophisticated style-based GANs on three medical imaging modalities and organs namely : cardiac cine-MRI, liver CT and RGB retina images. GANs were trained on well-known and widely utilized datasets from which their FID score were computed to measure the visual acuity of their generated images. We further tested their usefulness by measuring the segmentation accuracy of a U-Net trained on these generated images. Results reveal that GANs are far from being equal as some are ill-suited for medical imaging applications while others are much better off. The top-performing GANs are capable of generating realistic-looking medical images by FID standards that can fool trained experts in a visual Turing test and comply to some metrics. However, segmentation results suggests that no GAN is capable of reproducing the full richness of a medical datasets.

Monte-Carlo diffusion simulations are a powerful tool for validating tissue microstructure models by generating synthetic diffusion-weighted magnetic resonance images (DW-MRI) in controlled environments. This is fundamental for understanding the link between micrometre-scale tissue properties and DW-MRI signals measured at the millimetre-scale, optimising acquisition protocols to target microstructure properties of interest, and exploring the robustness and accuracy of estimation methods. However, accurate simulations require substrates that reflect the main microstructural features of the studied tissue. To address this challenge, we introduce a novel computational workflow, CACTUS (Computational Axonal Configurator for Tailored and Ultradense Substrates), for generating synthetic white matter substrates. Our approach allows constructing substrates with higher packing density than existing methods, up to 95 % intra-axonal volume fraction, and larger voxel sizes of up to (500um) 3 with rich fibre complexity. CACTUS generates bundles with angular dispersion, bundle crossings, and variations along the fibres of their inner and outer radii and g-ratio. We achieve this by introducing a novel global cost function and a fibre radial growth approach that allows substrates to match predefined targeted characteristics and mirror those reported in histological studies. CACTUS improves the development of complex synthetic substrates, paving the way for future applications in microstructure imaging.

This paper presents an overview of the second edition of the HEad and neCK TumOR (HECKTOR) challenge, organized as a satellite event of the 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2021. The challenge is composed of three tasks related to the automatic analysis of PET/CT images for patients with Head and Neck cancer (H&N), focusing on the oropharynx region. Task 1 is the automatic segmentation of H&N primary Gross Tumor Volume (GTVt) in FDG-PET/CT images. Task 2 is the automatic prediction of Progression Free Survival (PFS) from the same FDG-PET/CT. Finally, Task 3 is the same as Task 2 with ground truth GTVt annotations provided to the participants. The data were collected from six centers for a total of 325 images, split into 224 training and 101 testing cases. The interest in the challenge was highlighted by the important participation with 103 registered teams and 448 result submissions. The best methods obtained a Dice Similarity Coefficient (DSC) of 0.7591 in the first task, and a Concordance index (C-index) of 0.7196 and 0.6978 in Tasks 2 and 3, respectively. In all tasks, simplicity of the approach was found to be key to ensure generalization performance. The comparison of the PFS prediction performance in Tasks 2 and 3 suggests that providing the GTVt contour was not crucial to achieve best results, which indicates that fully automatic methods can be used. This potentially obviates the need for GTVt contouring, opening avenues for reproducible and large scale radiomics studies including thousands potential subjects.

AI systems often fail to deliver reliable predictions across all inputs, prompting the need for hybrid human-AI decision-making. Existing Learning to Defer (L2D) approaches address this by training deferral models, but these are sensitive to changes in expert composition and require significant retraining if experts change. We propose a training-free, model- and expert-agnostic framework for expert deferral based on conformal prediction. Our method uses the prediction set generated by a conformal predictor to identify label-specific uncertainty and selects the most discriminative expert using a segregativity criterion, measuring how well an expert distinguishes between the remaining plausible labels. Experiments on CIFAR10-H and ImageNet16-H show that our method consistently outperforms both the standalone model and the strongest expert, with accuracies attaining $99.57\pm0.10\%$ and $99.40\pm0.52\%$, while reducing expert workload by up to a factor of $11$. The method remains robust under degraded expert performance and shows a gradual performance drop in low-information settings. These results suggest a scalable, retraining-free alternative to L2D for real-world human-AI collaboration.

Common Representation Learning (CRL), wherein different descriptions (or views) of the data are embedded in a common subspace, is receiving a lot of attention recently. Two popular paradigms here are Canonical Correlation Analysis (CCA) based approaches and Autoencoder (AE) based approaches. CCA based approaches learn a joint representation by maximizing correlation of the views when projected to the common subspace. AE based methods learn a common representation by minimizing the error of reconstructing the two views. Each of these approaches has its own advantages and disadvantages. For example, while CCA based approaches outperform AE based approaches for the task of transfer learning, they are not as scalable as the latter. In this work we propose an AE based approach called Correlational Neural Network (CorrNet), that explicitly maximizes correlation among the views when projected to the common subspace. Through a series of experiments, we demonstrate that the proposed CorrNet is better than the above mentioned approaches with respect to its ability to learn correlated common representations. Further, we employ CorrNet for several cross language tasks and show that the representations learned using CorrNet perform better than the ones learned using other state of the art approaches.

Tractography algorithms leverage diffusion MRI to reconstruct the fibrous architecture of the brain's white matter. Among machine learning approaches, reinforcement learning (RL) has emerged as a promising framework for tractography, outperforming traditional methods in several key aspects. TractOracle-RL, a recent RL-based approach, reduces false positives by incorporating anatomical priors into the training process via a reward-based mechanism. In this paper, we investigate four extensions of the original TractOracle-RL framework by integrating recent advances in RL, and we evaluate their performance across five diverse diffusion MRI datasets. Results demonstrate that combining an oracle with the RL framework consistently leads to robust and reliable tractography, regardless of the specific method or dataset used. We also introduce a novel RL training scheme called Iterative Reward Training (IRT), inspired by the Reinforcement Learning from Human Feedback (RLHF) paradigm. Instead of relying on human input, IRT leverages bundle filtering methods to iteratively refine the oracle's guidance throughout training. Experimental results show that RL methods trained with oracle feedback significantly outperform widely used tractography techniques in terms of accuracy and anatomical validity.

Recent publications have shown that the segmentation accuracy of modern-day convolutional neural networks (CNN) applied on cardiac MRI can reach the inter-expert variability, a great achievement in this area of research. However, despite these successes, CNNs still produce anatomically inaccurate segmentations as they provide no guarantee on the anatomical plausibility of their outcome, even when using a shape prior. In this paper, we propose a cardiac MRI segmentation method which always produces anatomically plausible results. At the core of the method is an adversarial variational autoencoder (aVAE) whose latent space encodes a smooth manifold on which lies a large spectrum of valid cardiac shapes. This aVAE is used to automatically warp anatomically inaccurate cardiac shapes towards a close but correct shape. Our method can accommodate any cardiac segmentation method and convert its anatomically implausible results to plausible ones without affecting its overall geometric and clinical metrics. With our method, CNNs can now produce results that are both within the inter-expert variability and always anatomically plausible.

In this paper, we lay the groundwork on the comparison of phylogenetic networks based on edge contractions and expansions as edit operations, as originally proposed by Robinson and Foulds to compare trees. We prove that these operations connect the space of all phylogenetic networks on the same set of leaves, even if we forbid contractions that create cycles. This allows to define an operational distance on this space, as the minimum number of contractions and expansions required to transform one network into another. We highlight the difference between this distance and the computation of the maximum common contraction between two networks. Given its ability to outline a common structure between them, which can provide valuable biological insights, we study the algorithmic aspects of the latter. We first prove that computing a maximum common contraction between two networks is NP-hard, even when the maximum degree, the size of the common contraction, or the number of leaves is bounded. We also provide lower bounds to the problem based on the Exponential-Time Hypothesis. Nonetheless, we do provide a polynomial-time algorithm for weakly-galled networks, a generalization of galled trees.

In this paper, we present RT-GCC-NMF: a real-time (RT), two-channel blind speech enhancement algorithm that combines the non-negative matrix factorization (NMF) dictionary learning algorithm with the generalized cross-correlation (GCC) spatial localization method. Using a pre-learned universal NMF dictionary, RT-GCC-NMF operates in a frame-by-frame fashion by associating individual dictionary atoms to target speech or background interference based on their estimated time-delay of arrivals (TDOA). We evaluate RT-GCC-NMF on two-channel mixtures of speech and real-world noise from the Signal Separation and Evaluation Campaign (SiSEC). We demonstrate that this approach generalizes to new speakers, acoustic environments, and recording setups from very little training data, and outperforms all but one of the algorithms from the SiSEC challenge in terms of overall Perceptual Evaluation methods for Audio Source Separation (PEASS) scores and compares favourably to the ideal binary mask baseline. Over a wide range of input SNRs, we show that this approach simultaneously improves the PEASS and signal to noise ratio (SNR)-based Blind Source Separation (BSS) Eval objective quality metrics as well as the short-time objective intelligibility (STOI) and extended STOI (ESTOI) objective speech intelligibility metrics. A flexible, soft masking function in the space of NMF activation coefficients offers real-time control of the trade-off between interference suppression and target speaker fidelity. Finally, we use an asymmetric short-time Fourier transform (STFT) to reduce the inherent algorithmic latency of RT-GCC-NMF from 64 ms to 2 ms with no loss in performance. We demonstrate that latencies within the tolerable range for hearing aids are possible on current hardware platforms.

We investigate the problem of discovering and modeling regime shifts in an ecosystem comprising multiple time series known as co-evolving time series. Regime shifts refer to the changing behaviors exhibited by series at different time intervals. Learning these changing behaviors is a key step toward time series forecasting. While advances have been made, existing methods suffer from one or more of the following shortcomings: (1) failure to take relationships between time series into consideration for discovering regimes in multiple time series; (2) lack of an effective approach that models time-dependent behaviors exhibited by series; (3) difficulties in handling data discontinuities which may be informative. Most of the existing methods are unable to handle all of these three issues in a unified framework. This, therefore, motivates our effort to devise a principled approach for modeling interactions and time-dependency in co-evolving time series. Specifically, we model an ecosystem of multiple time series by summarizing the heavy ensemble of time series into a lighter and more meaningful structure called a \textit{mapping grid}. By using the mapping grid, our model first learns time series behavioral dependencies through a dynamic network representation, then learns the regime transition mechanism via a full time-dependent Cox regression model. The originality of our approach lies in modeling interactions between time series in regime identification and in modeling time-dependent regime transition probabilities, usually assumed to be static in existing work.

In this work, we study the problem of computing a maximum common contraction of two vertex-labeled graphs, i.e. how to make them identical by contracting as little edges as possible in the two graphs. We study the problem from a parameterized complexity point of view, using parameters such as the maximum degree, the degeneracy, the clique-width or treewidth of the input graphs as well as the number of allowed contractions. We put this complexity in perspective with that of the labeled contractibility problem, i.e determining whether a labeled graph is a contraction of another. Surprisingly, our results indicate very little difference between these problems in terms of parameterized complexity status. We only prove their status to differ when parameterizing by both the degeneracy and the number of allowed contractions, showing W[1]-hardness of the maximum common contraction problem in this case, whereas the contractibility problem is FPT.

The Massive Internet of Things (MIoT) envisions an interconnected ecosystem of billions of devices, fundamentally transforming diverse sectors such as healthcare, smart cities, transportation, agriculture, and energy management. However, the vast scale of MIoT introduces significant challenges, including network scalability, efficient data management, energy conservation, and robust security mechanisms. This paper presents a thorough review of existing and emerging MIoT technologies designed to address these challenges, including Low-Power Wide-Area Networks (LPWAN), 5G/6G capabilities, edge and fog computing architectures, and hybrid access methodologies. We further investigate advanced strategies such as AI-driven resource allocation, federated learning for privacy-preserving analytics, and decentralized security frameworks using blockchain. Additionally, we analyze sustainable practices, emphasizing energy harvesting and integrating green technologies to reduce environmental impact. Through extensive comparative analysis, this study identifies critical innovations and architectural adaptations required to support efficient, resilient, and scalable MIoT deployments. Key insights include the role of network slicing and intelligent resource management for scalability, adaptive protocols for real-time data handling, and lightweight AI models suited to the constraints of MIoT devices. This research ultimately contributes to a deeper understanding of how MIoT systems can evolve to meet the growing demand for seamless, reliable connectivity while prioritizing sustainability, security, and performance across diverse applications. Our findings serve as a roadmap for future advancements, underscoring the potential of MIoT to support a globally interconnected, intelligent infrastructure.

Federated Learning (FL) is a distributed machine learning technique, where each device contributes to the learning model by independently computing the gradient based on its local training data. It has recently become a hot research topic, as it promises several benefits related to data privacy and scalability. However, implementing FL at the network edge is challenging due to system and data heterogeneity and resources constraints. In this article, we examine the existing challenges and trade-offs in Federated Edge Learning (FEEL). The design of FEEL algorithms for resources-efficient learning raises several challenges. These challenges are essentially related to the multidisciplinary nature of the problem. As the data is the key component of the learning, this article advocates a new set of considerations for data characteristics in wireless scheduling algorithms in FEEL. Hence, we propose a general framework for the data-aware scheduling as a guideline for future research directions. We also discuss the main axes and requirements for data evaluation and some exploitable techniques and metrics.

We recently published a deep learning study on the potential of encoder-decoder networks for the segmentation of the 2D CAMUS ultrasound dataset. We propose in this abstract an extension of the evaluation criteria to anatomical assessment, as traditional geometric and clinical metrics in cardiac segmentation do not take into account the anatomical correctness of the predicted shapes. The completed study sheds a new light on the ranking of models.

We propose to consider a mutual incidence matrix $M$ of two balanced incomplete block designs built on the same finite set. In the simplest case, this matrix reduces to the standard incidence matrix of one block design. We find all eigenvalues of the matrices $MM^T$ and $M^TM$ and their eigenspaces.

We study the quasinormal modes (QNMs) of a charged scalar field on a Reissner-Nordstr\"{o}m-anti-de Sitter (RN-AdS$_{5}$) black hole in the small radius limit by using the isomonodromic method. We also derive the low-temperature expansion of the fundamental QNM frequency. Finally, we provide numerical evidence that instabilities appear in the small radius limit for large values of the charge of the scalar field.

This study presents the design, fabrication, and test of a micro accelerometer with intrinsic processing capabilities, that integrates the functions of sensing and computing in the same MEMS. The device consists of an inertial mass electrostatically coupled to an oscillating beam through a gap of 8 {\mu}m. The motion of the inertial mass modulates an AC electrostatic field that drives the beam in its non-linear regime. This non-linearity is used to implement machine learning in the mechanical domain, using reservoir computing with delayed feedback to process the acceleration information provided by the inertial mass. The device is microfabricated on a silicon-on-insulator substrate using conventional MEMS processes. Dynamic characterization showed good accelerometer functionalities, with an inertial mass sensitivity on the order of 100 mV/g from 250 to 1300 Hz and a natural frequency of 1.7 kHz. In order to test the device computing capabilities, two different machine learning benchmarks were implemented, with the inputs fed to the device as accelerations. The neuromorphic MEMS accelerometer was able to accurately emulate non-linear autoregressive moving average models and compute the parity of random bit streams. These results were obtained in a test system with a non-trivial transfer function, showing a robustness that is well-suited to anticipated applications.