Glioblastoma exhibits diverse, infiltrative, and patient-specific growth patterns that are only partially visible on routine MRI, making it difficult to reliably assess true tumor extent and personalize treatment planning and follow-up. We present a biophysically-conditioned generative framework that synthesizes biologically realistic 3D brain MRI volumes from estimated, spatially continuous tumor-concentration fields. Our approach combines a generative model with tumor-infiltration maps that can be propagated through time using a biophysical growth model, enabling fine-grained control over tumor shape and growth while preserving patient anatomy. This enables us to synthesize consistent tumor growth trajectories directly in the space of real patients, providing interpretable, controllable estimation of tumor infiltration and progression beyond what is explicitly observed in imaging. We evaluate the framework on longitudinal glioblastoma cases and demonstrate that it can generate temporally coherent sequences with realistic changes in tumor appearance and surrounding tissue response. These results suggest that integrating mechanistic tumor growth priors with modern generative modeling can provide a practical tool for patient-specific progression visualization and for generating controlled synthetic data to support downstream neuro-oncology workflows. In longitudinal extrapolation, we achieve a consistent 75% Dice overlap with the biophysical model while maintaining a constant PSNR of 25 in the surrounding tissue. Our code is available at: https://github.com/valentin-biller/lgm.git

In this paper, considering that mosquitoes have a much faster reproductive rate than humans, a two-time scaled dengue model is formulated by using time scale transformation. By proving the existence of an invariant measure with exponentially ergodic property in the mosquito dynamic equations, we further show that the human subsystem strongly converges to the solution of the averaged equations. Moreover, spraying mosquito insecticides and treating infected persons are introduced as control measures, and the optimal control model of dengue is developed. Based on the convex perturbation method, the first order necessary conditions for optimal control are derived. Numerical simulations are provided to explain and supplement the theoretical result obtained.

Large-scale multi-modal MRI datasets impose substantial storage and I/O costs, limiting the training of 3D generative models on commodity infrastructure. While lossy compression is known to preserve accuracy for discriminative segmentation networks, its effect on generative models, which must learn the full data distribution rather than a decision boundary, is unexplored. We study whether standard image codecs can effectively compress semantically rich brain tumor MRI while preserving the fidelity required to train and deploy a 3D MRI generative model. Each 3D volume is compressed with JPEG2000 or a near-lossless JPEG-LS pipeline. Next, a Wavelet Flow Matching model, conditioned on BraTS image sequences (T1n, T1c, T2, T2f), is trained on compressed data, and the resulting models are evaluated on the validation set. At a 20:1 compression ratio, synthesis quality is statistically equivalent to a model trained on uncompressed data within a pre-specified margin (ΔPSNR <1,dB, ΔSSIM <0.02; paired TOST p=[[p]]): mean PSNR is 27.3,dB vs. 27.0,dB and mean SSIM is 0.95 vs. 0.96 across modalities. Our results indicate that JPEG2000 compression is a practical step toward scalable 3D MRI generative modeling without degrading synthesis quality. The codebase is available at https://github.com/lisafis/MRIComp4Flow .

Vision foundation models like DINOv2 demonstrate remarkable potential in medical imaging despite their origin in natural image domains. However, their design inherently works best for uni-modal image analysis, limiting their effectiveness for multi-modal imaging tasks that are common in many medical fields, such as neurology and oncology. While supervised models perform well in this setting, they fail to leverage unlabeled datasets and struggle with missing modalities—a frequent challenge in clinical settings. To bridge these gaps, we introduce MM-DINOv2, a novel and efficient framework that adapts the pre-trained vision foundation model DINOv2 for multi-modal medical imaging. Our approach incorporates multi-modal patch embeddings, enabling vision foundation models to effectively process multi-modal imaging data. To address missing modalities, we employ full-modality masking, which encourages the model to learn robust cross-modality relationships. Furthermore, we leverage semi-supervised learning to harness large unlabeled datasets, enhancing both the accuracy and reliability of medical predictions. We demonstrate our approach on glioma subtype classification from multi-sequence brain MRI, achieving a Matthews Correlation Coefficient (MCC) of 0.6 on an external test set, surpassing state-of-the-art supervised approaches by +11.1 https://github.com/daniel-scholz/mm-dinov2 ).

This paper investigates dynamical behaviors of a population model in a polluted environment with non-continuous threshold harvesting. The model is characterized by time delay and reaction-diffusion. Firstly, we present sufficient conditions for the local asymptotic stability of the regular equilibrium and the existence of Hopf bifurcation through an analysis of the associated characteristic equation. Then we take a further step to determine the direction of Hopf bifurcation and the stability of bifurcated periodic solutions by utilizing normal form theory and center manifold reduction. Moreover, the sliding domain and its dynamics are discussed by employing the Utkin equivalent control method. Finally, several numerical simulations are conducted to support our theoretical analysis results, and the boundary node bifurcations are also illustrated. (c) 2024 Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

The growth rates of avian and human populations are remarkably different on time scales pertaining to different biological characteristics, and the spread of avian influenza virus is often affected by external environmental perturbations and spatial heterogeneity. In this paper, we formulate a stochastic partial differential equation for avian influenza model with fast and slow time scales. The existence and uniqueness of the positive solutions are proved by using the Yamada–Watanabe approximation principle. The existence and uniqueness of the invariant measure within avian sub-system are demonstrated by applying the averaging principle and ergodic property. By using relaxed control expression along with martingale technique, the human subsystem’s weak convergence is proved, which means that the avian subsystem weakly converges to the solutions of a corresponding averaged equation. Furthermore, by implementing poultry culling and medical treatment of infected cases as control strategies, the near-optimality of control measures of the multiscale stochastic system is obtained by establishing the near-optimality of control measures of the average equations for human subsystem. Finally, numerical simulations are given to prove and understand our theoretical results.

BACKGROUND:Glioblastomas are functionally integrated into their peritumoral neural environment, and the dynamic functional interaction can be analyzed using network theory, providing insights into the tumor-brain interface. We investigated the peritumoral network connectedness of glioblastomas, revealing its association with distinct epigenetic signatures, its influence on survival, and its susceptibility to modification through surgical treatment. METHODS:Resting-state fMRI was performed on 48 glioblastoma patients. Tumor lesions were segmented, and networks were constructed at 10 mm and 40 mm distances from the tumor margin. These networks were mirrored to the healthy hemisphere to compare lesional and contralesional networks. The difference between lesional and contralesional mean degree centrality was calculated to assess the peritumoral network connectedness. Its correlation with epigenetic signatures and effect on overall survival were analyzed. Surgery-induced changes in the peritumoral network connectedness were evaluated in 7 patients with follow-up data. RESULTS:Mean degree centrality was significantly higher in the lesional compared to the contralesional network (P = .032), indicating a tumor-induced effect on its local environment and reflecting high peritumoral network connectedness. Glioblastomas with a neural high epigenetic signature exhibited increased peritumoral network connectedness (P = .010), which was associated with decreased survival (P = .036). Postoperative peritumoral network connectedness tended to decrease, suggesting that surgical resection disrupts the functional communication between the tumor and its peritumoral environment. CONCLUSIONS:The role of network features in predicting patient survival suggests their clinical relevance as imaging biomarkers for assessing personalized treatment strategies, which may include targeting crucial nodes for disconnection or even neuromodulation of neural circuits.

We analyze structural and functional connectivity graphs for both healthy and glioma subjects by clustering-based model reduction to obtain a detailed understanding of these complex graphs. The reduced-order network is obtained by dividing the connected brain regions or nodes into disjoint clusters. Further, each cluster is replaced by a single new node in a reduced-size graph network. Several partitions are analyzed and evaluated based on model reduction errors.

In this paper, a piecewise smooth dengue model with threshold policy is developed, and the impacts of different control measures on the spread of dengue fever are also investigated. The number of dengue infections is used as a threshold level to determine whether to implement control, and control measures are triggered only if the number of infected individuals exceeds . Using the Routh–Hurwitz criterion, the dynamic behaviors of the free system and control system are studied. Then, the existence of the sliding mode is verified, and the sliding dynamics are analyzed by using the Utkin equivalent control method. It is shown that model solutions eventually converge to one of two endemic equilibria or the sliding equilibrium depending on . In addition, by using Ekeland's principle and maximum principle, the sufficient and necessary conditions for near‐optimal controls of this model are obtained. Finally, numerical simulations are carried out to explain and supplement the theoretical results.

A class of time-varying delay impulsive reaction–diffusion tree–grass–water–nitrogen system driven by Lévy jump process is considered. First, we prove the existence and uniqueness of the global positive solution of the model by constructing the Lyapunov function. Secondly, several sufficient conditions for finite-time stability are given by using comparison theorem and mean impulse interval method. Finally, numerical simulations are carried out to verify the effectiveness of the theoretical analysis.

Normal and aberrant cognitive functions are the result of the dynamic interplay between large-scale neural circuits. Describing the nature of these interactions has been a challenging task yet important for neurodegenerative disease evolution. Fusing modern dynamic graph network theory techniques and control theory applied on complex brain networks creates a new framework for neurodegenerative disease research by determining disease evolution at the subject level, facilitating a predictive treatment response and revealing key mechanisms responsible for disease alterations. It has been shown that two types of controllability—the average and the modal controllability—are relevant for the mechanistic explanation of how the brain navigates between cognitive states. The average controllability favors highly connected areas which move the brain to easily reachable states, while the modal controllability favors weakly connected areas representative for difficult-to-reach states. We propose two different techniques to achieve these two types of controllability: a centrality measure based on a sensitivity analysis of the Laplacian matrix is employed to determine the average controllability, while graph distances form the basis of the modal controllability. The concepts of “choosing the best driver set” and “graph distances” are applied to measure the average controllability and the modal controllability, respectively. Based on these new techniques, we obtain important disease descriptors that visualize alterations in the disease trajectory by revealing densely connected hubs or sparser areas. Our results suggest that these two techniques can accurately describe the different node roles in controlling trajectories of brain networks.

Brain connectivity is usually analyzed based on graph theory and pinning control theory. Previous studies suggested that the topological properties of structural and functional networks for brain networks may be altered in association with neurodegnerative diseases. To better understand and characterize these alterations, we introduce a new approach - robustness of network controllability to evaluate network robustness, and identify the critical nodes, whose removals maximally destroys the network's functionality. These alterations are due to external or internal changes in the network. Understanding and describing these interactions at the level of large-scale brain circuitry may be a significant step towards unraveling dementia disease evolution. In this study, we analyze structural and functional brain networks for healthy controls, MCI and AD patients such that we reveal the connection between network robustness and architecture and the differences between patients' groups. We determine the critical and driver nodes of these networks as the key components for robustness of network controllability. Our results suggest that healthy controls for both functional and structural connectivity have more critical nodes than AD and MCI networks, and that these critical nodes appear clustered in almost all networks. Our findings provide useful information for determining disease evolution in dementia under the aspects of controllability and robustness.

Brain networks can be naturally divided into clusters or communities where the cluster's nodes dynamics have similar trajectories in phase space. This process is known as synchronization, and represents characteristics of intragroup features and not between groups. Fractional calculus represents a generalization of ordinary differentiation and integration to arbitrary non-integer order, and can be thought of as a smooth interpolation between different orders of differentiation/integration, providing the ability to probe the system from many different viewpoints of the dynamics. Fractional calculus has been explored as an excellent tool for the description of memory in many processes and may be more accurate for modeling brain processes than traditional integer-order ones. We apply the concept of cluster synchronization in fractional-order structural brain networks ranging from healthy controls to Alzheimer's disease subjects and determine whether cluster synchronization can be achieved in these networks. We observe the existence of a hypersynchronization only in AD structural networks and consider that this could represent an excellent non-invasive biomarker for tracking the disease evolution and decide upon therapeutic interventions.

The mutational status of the isocitrate dehydrogenase (IDH) gene plays a key role in the treatment of glioma patients because it is known to affect energy metabolism pathways relevant to glioma. Physio-metabolic magnetic resonance imaging (MRI) enables the non-invasive analysis of oxygen metabolism and tissue hypoxia as well as associated neovascularization and microvascular architecture. However, evaluating such complex neuroimaging data requires computational support. Traditional machine learning algorithms and simple deep learning models were trained with radiomic features from clinical MRI (cMRI) or physio-metabolic MRI data. A total of 215 patients (first center: 166 participants + 16 participants for independent internal testing of the algorithms versus second site: 33 participants for independent external testing) were enrolled using two different physio-metabolic MRI protocols. The algorithms trained with physio-metabolic data demonstrated the best classification performance in independent internal testing: precision, 91.7%; accuracy, 87.5%; area under the receiver operating curve (AUROC), 0.979. In external testing, traditional machine learning models trained with cMRI data exhibited the best IDH classification results: precision, 84.9%; accuracy, 81.8%; and AUROC, 0.879. The poor performance for the physio-metabolic MRI approach appears to be explainable by site-dependent differences in data acquisition methodologies. The physio-metabolic MRI approach potentially supports reliable classification of IDH gene status in the presurgical stage of glioma patients. However, non-standardized protocols limit the level of evidence and underlie the need for a reproducible framework of data acquisition techniques.

A variety of deep learning approaches have been proposed to automatically classify Alzheimer’s disease (AD) from medical images. However, common approaches such as traditional convolutional neural networks (CNNs), lack interpretability and are prone to overfitting when trained on small datasets. As an alternative, significantly less work has explored applying deep learning approaches to region-based features that are commonly attained from atlas partitions of known regions of interest (ROI). In this work, we combine CNNs with graph neural networks (GNNs) to jointly learn an adjacency matrix of connectivity’s between ROIs as a prior for learning meaningful features for AD prediction. We apply our method to the ADNI dataset and systematically inspect the different intermediate layers of our network using t-SNE projections that show strong separation on out-of-sample data. Finally, we show that the edge probabilities alone are sufficient to reach high classification accuracy by training a secondary random forest classifier on the adjacency matrices outputted from our network and illustrate the interpretability properties of the graphs by visualizing the feature importance for all edges.

Artificial intelligence (AI) is considered one of the core technologies of the Fourth Industrial Revolution that is currently taking place [...].

Previous studies suggest that the topological properties of structural and functional neural networks in glioma patients are altered beyond the tumor location. These alterations are due to the dynamic interactions with large-scale neural circuits. Understanding and describing these interactions may be an important step towards deciphering glioma disease evolution. In this study, we analyze structural and functional brain networks in terms of determining the correlation between network robustness and topological features regarding the default-mode network (DMN), comparing prognostically differing patient groups to healthy controls. We determine the driver nodes of these networks, which are receptive to outside signals, and the critical nodes as the most important elements for controllability since their removal will dramatically affect network controllability. Our results suggest that network controllability and robustness of the DMN is decreased in glioma patients. We found losses of driver and critical nodes in patients, especially in the prognostically less favorable IDH wildtype (IDHwt) patients, which might reflect lesion-induced network disintegration. On the other hand, topological shifts of driver and critical nodes, and even increases in the number of critical nodes, were observed mainly in IDH mutated (IDHmut) patients, which might relate to varying degrees of network plasticity accompanying the chronic disease course in some of the patients, depending on tumor growth dynamics. We hereby implement a novel approach for further exploring disease evolution in brain cancer under the aspects of neural network controllability and robustness in glioma patients.

The precise initial characterization of contrast-enhancing brain tumors has significant consequences for clinical outcomes. Various novel neuroimaging methods have been developed to increase the specificity of conventional magnetic resonance imaging (cMRI) but also the increased complexity of data analysis. Artificial intelligence offers new options to manage this challenge in clinical settings. Here, we investigated whether multiclass machine learning (ML) algorithms applied to a high-dimensional panel of radiomic features from advanced MRI (advMRI) and physiological MRI (phyMRI; thus, radiophysiomics) could reliably classify contrast-enhancing brain tumors. The recently developed phyMRI technique enables the quantitative assessment of microvascular architecture, neovascularization, oxygen metabolism, and tissue hypoxia. A training cohort of 167 patients suffering from one of the five most common brain tumor entities (glioblastoma, anaplastic glioma, meningioma, primary CNS lymphoma, or brain metastasis), combined with nine common ML algorithms, was used to develop overall 135 classifiers. Multiclass classification performance was investigated using tenfold cross-validation and an independent test cohort. Adaptive boosting and random forest in combination with advMRI and phyMRI data were superior to human reading in accuracy (0.875 vs. 0.850), precision (0.862 vs. 0.798), F-score (0.774 vs. 0.740), AUROC (0.886 vs. 0.813), and classification error (5 vs. 6). The radiologists, however, showed a higher sensitivity (0.767 vs. 0.750) and specificity (0.925 vs. 0.902). We demonstrated that ML-based radiophysiomics could be helpful in the clinical routine diagnosis of contrast-enhancing brain tumors; however, a high expenditure of time and work for data preprocessing requires the inclusion of deep neural networks.

We study the optimal control of the mean and variance of the network state vector. We develop an algorithm that uses projected gradient descent to optimize the control input placement, subject to constraints on the state that must be achieved at a given time threshold; seeking to design an input that moves the moment at minimum cost. First, we solve the state-selection problem for a number of variants of the first and second moment, and find solutions related to the eigenvalues of the systems' Gramian matrices. We then nest this state selection into projected gradient descent to design optimal inputs.