Video-to-Speech (VTS) generation aims to synthesize speech from a silent video without auditory signals. However, existing VTS methods disregard the hierarchical nature of speech, which spans coarse speaker-aware semantics to fine-grained prosodic details. This oversight hinders direct alignment between visual and speech features at specific hierarchical levels during property matching. In this paper, leveraging the hierarchical structure of Residual Vector Quantization (RVQ)-based codec, we propose HiCoDiT, a novel Hierarchical Codec Diffusion Transformer that exploits the inherent hierarchy of discrete speech tokens to achieve strong audio-visual alignment. Specifically, since lower-level tokens encode coarse speaker-aware semantics and higher-level tokens capture fine-grained prosody, HiCoDiT employs low-level and high-level blocks to generate tokens at different levels. The low-level blocks condition on lip-synchronized motion and facial identity to capture speaker-aware content, while the high-level blocks use facial expression to modulate prosodic dynamics. Finally, to enable more effective coarse-to-fine conditioning, we propose a dual-scale adaptive instance layer normalization that jointly captures global vocal style through channel-wise normalization and local prosody dynamics through temporal-wise normalization. Extensive experiments demonstrate that HiCoDiT outperforms baselines in fidelity and expressiveness, highlighting the potential of discrete modelling for VTS. The code and speech demo are both available at https://github.com/Jiaxin-Ye/HiCoDiT.

Advances in multimodal large language models (MLLMs) are extending radiological artificial intelligence (AI) beyond task-specific image analysis toward multimodal understanding and reasoning. Volumetric radiology, however, presents a fundamental representational mismatch: clinical interpretation often requires full-volume spatial context and acquisition-dependent quantitative information, whereas current MLLMs are commonly conditioned on selected two-dimensional (2D) images, compressed visual representations, or report-derived text. Reliable volumetric radiology AI therefore requires representations that preserve task-relevant three-dimensional (3D) information and systems that can access, verify, and integrate this information across clinical workflows. In this Review, we examine more than 200 publications through July 2026. We organize the literature around volumetric representation and multimodal understanding at the model level, agentic orchestration at the system level, and their links to clinical applications and evaluation. We review volumetric foundation models, language alignment and compression strategies, and agentic systems that extend MLLMs through planning, tools, memory, and workflow interaction. We distinguish settings in which selected 2D views or report-mediated reasoning may suffice from those that warrant native volumetric modeling. We also introduce a Claim-Design-Validation framework to assess whether technical, workflow, and clinical claims are matched by appropriate design and validation. Across the literature, native volumetric modeling and agentic capabilities depend on the spatial, quantitative, contextual, and workflow requirements of the intended task. Clinical credibility requires faithful volumetric representation, traceable system behavior, claim-aligned validation, and clearly defined human oversight in realistic workflows.

Free-text promptable 3D medical image segmentation offers an intuitive and clinically flexible interaction paradigm. However, current methods are highly sensitive to linguistic variability: minor changes in phrasing can cause substantial performance degradation despite identical clinical intent. Existing approaches attempt to improve robustness through stronger vision-language fusion or larger vocabularies, yet they lack mechanisms to consistently align ambiguous free-form expressions with anatomically grounded representations. We propose Skill-Evolving grounded Reasoning (SEER), a novel framework for free-text promptable 3D medical image segmentation that explicitly bridges linguistic variability and anatomical precision through a reasoning-driven design. First, we curate the SEER-Trace dataset, which pairs raw clinical requests with image-grounded, skill-tagged reasoning traces, establishing a reproducible benchmark. Second, SEER constructs an evidence-aligned target representation via a vision-language reasoning chain that verifies clinical intent against image-derived anatomical evidence, thereby enforcing semantic consistency before voxel-level decoding. Third, we introduce SEER-Loop, a dynamic skill-evolving strategy that distills high-reward reasoning trajectories into reusable skill artifacts and progressively integrates them into subsequent inference, enabling structured self-refinement and improved robustness to diverse linguistic expressions. Extensive experiments demonstrate superior performance of SEER over state-of-the-art baselines. Under linguistic perturbations, SEER reduces performance variance by 81.94

Alzheimer's disease (AD) is a progressive neurodegenerative disorder for which MRI and PET provide complementary structural and molecular information. Yet PET remains costly and often unavailable, motivating MRI-only diagnostic systems that still benefit from multimodal supervision. Existing methods either synthesize PET from MRI or limit cross-modal learning to low-dimensional spaces, underutilizing MRI-PET complementarity and leading to limited robustness and generalizability. To address these challenges, we introduce E2AD, an Enhanced and Explainable AD detection framework that leverages anatomy- and relation-aware cross-modal knowledge distillation (KD). Using paired MRI-PET data during training but only MRI at inference, E2AD augments traditional logit-based KD through two synergistic components: (1) anatomy-aware distillation that transfers within-subject anatomical dependencies through an anatomical Mixture-of-Mappers, yielding spatially meaningful and clinically traceable cues; and (2) relation-aware distillation that promotes stable between-subject structural relations through generalizable pairwise alignment, yielding a representation space with better cross-cohort generalization. To enhance clinical utility, we further introduce a tailored multi-agent workflow that translates E2AD's anatomical attention into structured, clinician-oriented MRI reports. Extensive experimental results on the internal ADNI cohort and two external cohorts (AIBL and NACC) demonstrate that E2AD outperforms state-of-the-art baselines, offering faster convergence, improved data efficiency, stronger cross-cohort generalization, and enhanced explainability. Source code is available at https://github.com/thibault-wch/E2AD-for-Alzheimer-disease.

Positron emission tomography (PET) plays a critical role in the early detection of Alzheimer's disease (AD) by revealing pathophysiological changes prior to clinical onset or neurodegeneration detectable by magnetic resonance imaging (MRI). However, the clinical deployment of multi-tracer PET is limited by high cost and restricted accessibility, especially in resource-limited settings. Here, we introduce MaM-DiT, a novel modality-aware manifold diffusion transformer that synthesizes high-quality, diagnostically informative multi-tracer PET images directly from MRI. By capturing spatiotemporal features across the amyloid–tau–neurodegeneration (A−T−N) spectrum, MaM-DiT enables efficient, non-invasive assessment of AD pathology. The model was trained on 869 paired MRI–PET scans from three clinical centers, visually validated on 106 samples from an independent center, and diagnostically evaluated on 1,207 subjects from the ADNI cohort. We comprehensively assessed the image quality and diagnostic utility of synthetic PETs and found that they exhibited high visual fidelity to real scans across tracers (SSIM > 0.87). Moreover, synthetic PETs enabled AD subtype stratification and improved MRI-along diagnostic performance, yielding gains of up to 13.8% in AUC-ROC and 12.5% in AUC-PR. These results highlight the potential of AI-synthesized multi-tracer PET as a cost-effective prescreening tool to triage at-risk individuals, supporting early detection and optimized resource allocation in diverse healthcare settings.

Diffusion transformers (DiTs) have emerged as a powerful architecture for high-fidelity image generation, yet the quadratic cost of self-attention poses a major scalability bottleneck. To address this, linear attention mechanisms have been adopted to reduce computational cost; unfortunately, the resulting linear diffusion transformers (LiTs) models often come at the expense of generative performance, frequently producing over-smoothed attention weights that limit expressiveness. In this work, we introduce Dynamic Differential Linear Attention (DyDiLA), a novel linear attention formulation that enhances the effectiveness of LiTs by mitigating the oversmoothing issue and improving generation quality. Specifically, the novelty of DyDiLA lies in three key designs: (i) dynamic projection module, which facilitates the decoupling of token representations by learning with dynamically assigned knowledge; (ii) dynamic measure kernel, which provides a better similarity measurement to capture fine-grained semantic distinctions between tokens by dynamically assigning kernel functions for token processing; and (iii) token differential operator, which enables more robust query-to-key retrieval by calculating the differences between the tokens and their corresponding information redundancy produced by dynamic measure kernel. To capitalize on DyDiLA, we introduce a refined LiT, termed DyDi-LiT, that systematically incorporates our advancements. Extensive experiments show that DyDi-LiT consistently outperforms current state-of-the-art (SOTA) models across multiple metrics, underscoring its strong practical potential.

Precise prognostic modeling of glioblastoma (GBM) under varying treatment interventions is essential for optimizing clinical outcomes. While generative AI has shown promise in simulating GBM evolution, existing methods typically treat interventions as static conditional inputs rather than dynamic decision variables. Consequently, they fail to capture the complex, reciprocal interplay between tumor evolution and treatment response. To bridge this gap, we present Brain-WM, a pioneering brain GBM world model that unifies next-step treatment prediction and future MRI generation, thereby capturing the co-evolutionary dynamics between tumor and treatment. Specifically, Brain-WM encodes spatiotemporal dynamics into a shared latent space for joint autoregressive treatment prediction and flow-based future MRI generation. Then, instead of a conventional monolithic framework, Brain-WM adopts a novel Y-shaped Mixture-of-Transformers (MoT) architecture. This design structurally disentangles heterogeneous objectives, successfully leveraging cross-task synergies while preventing feature collapse. Finally, a synergistic multi-timepoint mask alignment objective explicitly anchors latent representations to anatomically grounded tumor structures and progression-aware semantics. Extensive validation on internal and external multi-institutional cohorts demonstrates the superiority of Brain-WM, achieving 91.5

Reconstructing dynamic visual experiences as videos from functional magnetic resonance imaging (fMRI) is pivotal for advancing the understanding of neural processes. However, current fMRI-to-video reconstruction methods are hindered by a semantic gap between noisy fMRI signals and the rich content of videos, stemming from a reliance on incomplete semantic embeddings that neither capture video-specific cues (e.g., actions) nor integrate prior knowledge. To this end, we draw inspiration from the dual-pathway processing mechanism in human brain and introduce CineNeuron, a novel hierarchical framework for semantically enhanced video reconstruction from fMRI signals with two synergistic stages. First, a bottom-up semantic enrichment stage maps fMRI signals to a rich embedding space that comprehensively captures textual semantics, image contents, action concepts, and object categories. Second, a top-down memory integration stage utilizes the proposed Mixture-of-Memories method to dynamically select relevant "memories" from previously seen data and fuse them with the fMRI embedding to refine the video reconstruction. Extensive experimental results on two fMRI-to-video benchmarks demonstrate that CineNeuron surpasses state-of-the-art methods across various metrics.

Molecular subtyping is essential for guiding systemic therapy in breast cancer but currently requires invasive biopsy. Conventional B-mode ultrasound offers rich anatomical information, yet lacks the functional dynamics needed to capture the comprehensive biology of tumors. Here, we present the first multimodal ultrasound spatiotemporal transformer, MUST-Sub, which integrates paired B-mode morphological features with contrast-enhanced ultrasound (CEUS) hemodynamic patterns to classify Luminal, HER2-enriched, and triple-negative subtypes. Training on a retrospective development cohort, and validated on internal, prospective, and multicenter external cohorts, MUST-Sub achieved macro-average areas under the receiver operating characteristic curve (AUCs) of 0.94, 0.90, and 0.92, respectively, and Luminal versus non-Luminal AUCs of 0.92, 0.88, and 0.91, outperforming B-mode-only deep learning baselines. MUST-Sub also produced interpretable quantitative biomarkers derived from spatiotemporal attention: the morphology-associated biomarker showed inverse correlations with tumor size (Spearman ρ = [- 0.34, - 0.23]; all p < . 05), while the hemodynamics-associated biomarker correlated positively with tumor size (ρ = [0.24, 0.32]; all p < . 05) and Ki-67 proliferation index (ρ = [0.21, 0.24]; all p < . 05). These results suggest that multimodal ultrasound with spatiotemporal modeling can serve as a promising adjunctive approach for non-invasive pre-biopsy molecular phenotyping of breast cancer.

Autoregressive modeling has driven major advances in multimodal AI, yet its application to medical imaging remains constrained by the absence of a unified image tokenizer that simultaneously preserves fine-grained anatomical structures and rich clinical semantics across heterogeneous modalities. Existing approaches jointly optimize image reconstruction and textual semantic objectives, relying on large-scale image-caption pairs and are prone to gradient interference. This is ill-suited for the medical domain where paired data are scarce and abundant unpaired images remain unexploited. This work identifies these issues in building unified medical image tokenizers, and introduces a principled two-stage training framework using visual representation as a bridge to address them. The propose visual representation alignment stage enables the utilization of large-scale unpaired medical images to ensure reconstruction fidelity and establish foundational semantics, alleviating the interference and better preparing for the second stage where fine-grained textual semantics are injected using image-text pairs. The resulting tokenizer, MedITok, is trained on over 33 million medical images spanning 9 modalities and 2 million image-text pairs. MedITok achieves state-of-the-art performance on 30+ benchmarks spanning 9 imaging modalities and 4 task families. It further enables autoregressive modeling for diagnostic and generative applications, serving as a scalable component for future multimodal models with unified synthesis and understanding capabilities in the medical domain. Project page: https://github.com/Masaaki-75/meditok

Low-dose computed tomography (LDCT) reduces radiation exposure but often degrades image quality, potentially compromising diagnostic accuracy. Existing deep learning-based denoising methods focus primarily on pixel-level mappings, overlooking the potential benefits of high-level semantic guidance. Recent advances in vision-language models (VLMs) suggest that language can serve as a powerful tool for capturing structured semantic information, offering new opportunities to improve LDCT reconstruction. In this paper, we introduce LangMamba, a Language-driven Mamba framework for LDCT denoising that leverages VLM-derived representations to enhance supervision from normal-dose CT (NDCT). LangMamba follows a two-stage learning strategy. First, we pre-train a Language-guided AutoEncoder (LangAE) that leverages frozen VLMs to map NDCT images into a semantic space enriched with anatomical information. Second, we synergize LangAE with two key components to guide LDCT denoising: Semantic-Enhanced Efficient Denoiser (SEED), which enhances NDCT-relevant local semantic while capturing global features with efficient Mamba mechanism, and Language-engaged Dual-space Alignment (LangDA) Loss, which ensures that denoised images align with NDCT in both perceptual and semantic spaces. Extensive experiments on two public datasets demonstrate that LangMamba outperforms conventional state-of-the-art methods, significantly improving detail preservation and visual fidelity. Remarkably, LangAE exhibits strong generalizability to unseen datasets, thereby reducing training costs. Furthermore, LangDA loss improves explainability by integrating language-guided insights into image reconstruction and offers a plug-and-play fashion. Our findings shed new light on the potential of language as a supervisory signal to advance LDCT denoising. The code is publicly available on https://github.com/hao1635/LangMamba.

While deep learning has significantly advanced medical image segmentation, most existing methods still struggle with handling complex anatomical regions. Cascaded or deep supervision-based approaches attempt to address this challenge through multi-scale feature learning but fail to establish sufficient inter-scale dependencies, as each scale relies solely on the features of the immediate predecessor. Towards this end, we propose the AutoRegressive Segmentation framework via next-scale mask prediction, termed AR-Seg, which progressively predicts the next-scale mask by explicitly modeling dependencies across all previous scales within a unified architecture. AR-Seg introduces three innovations: (1) a multi-scale mask autoencoder that quantizes the mask into multi-scale token maps to capture hierarchical anatomical structures, (2) a next-scale autoregressive mechanism that progressively predicts next-scale masks to enable sufficient inter-scale dependencies, and (3) a consensus-aggregation strategy that combines multiple sampled results to generate a more accurate mask, further improving segmentation robustness. Extensive experimental results on two benchmark datasets with different modalities demonstrate that AR-Seg outperforms state-of-the-art methods while explicitly visualizing the intermediate coarse-to-fine segmentation process. Source code is made available at https://github.com/takimailto/AR-Seg

Medical image segmentation has been significantly advanced with the rapid development of deep learning (DL) techniques. Existing DL-based segmentation models are typically discriminative; i.e ., they aim to learn a mapping from the input image to segmentation masks. However, these discriminative methods neglect the underlying data distribution and intrinsic class characteristics, suffering from unstable feature space. In this work, we propose to complement discriminative segmentation methods with the knowledge of underlying data distribution from generative models. To that end, we propose a novel hybrid diffusion framework for medical image segmentation, termed HiDiff, which can synergize the strengths of existing discriminative segmentation models and new generative diffusion models. HiDiff comprises two key components: discriminative segmentor and diffusion refiner. First, we utilize any conventional trained segmentation models as discriminative segmentor, which can provide a segmentation mask prior for diffusion refiner. Second, we propose a novel binary Bernoulli diffusion model (BBDM) as the diffusion refiner, which can effectively, efficiently, and interactively refine the segmentation mask by modeling the underlying data distribution. Third, we train the segmentor and BBDM in an alternate-collaborative manner to mutually boost each other. Extensive experimental results on abdomen organ, brain tumor, polyps, and retinal vessels segmentation datasets, covering four widely-used modalities, demonstrate the superior performance of HiDiff over existing medical segmentation algorithms, including the state-of-the-art transformer- and diffusion-based ones. In addition, HiDiff excels at segmenting small objects and generalizing to new datasets. Source codes are made available at https://github.com/takimailto/HiDiff.

While various deep learning methods were proposed for low-dose computed tomography (CT) denoising, they often suffer from over-smoothing, blurring, and lack of explainability. To alleviate these issues, we propose a plug-and-play LanguageEngaged Dual-space Alignment loss (LEDA) to optimize low-dose CT denoising models. Our idea is to leverage large language models (LLMs) to align denoised CT and normal-dose CT images in both the continuous perceptual space and discrete semantic space, which is the first LLM-based scheme for lowdose CT denoising. LEDA involves two steps: the first is to pretrain an LLM-guided CT autoencoder, which can encode a CT image into continuous high-level features and quantize them into a token space to produce semantic tokens derived from the LLM's vocabulary; and the second is to minimize the discrepancy between the denoised CT images and normaldose CT in terms of both encoded high-level features and quantized token embeddings derived by the LLM-guided CT autoencoder. Extensive experimental results demonstrate that our LEDA can enhance existing denoising models in terms of quantitative metrics and qualitative evaluation, and also provide explainability through language-level image understanding. The code is publicly available at https://github.com/hao1635/LEDA.

Mild cognitive impairment (MCI) is often at high risk of progression toAlzheimer's disease (AD). Existing works to identify the progressive MCI (pMCI)typically require MCI subtype labels, pMCI vs. stable MCI (sMCI), determined bywhether or not an MCI patient will progress to AD after a long follow-up.However, prospectively acquiring MCI subtype data is time-consuming andresource-intensive; the resultant small datasets could lead to severeoverfitting and difficulty in extracting discriminative information. Inspiredby that various longitudinal biomarkers and cognitive measurements present anordinal pathway on AD progression, we propose a novel Hybrid-granularityOrdinal PrototypE learning (HOPE) method to characterize AD ordinal progressionfor MCI progression prediction. First, HOPE learns an ordinal metric space thatenables progression prediction by prototype comparison. Second, HOPE leveragesa novel hybrid-granularity ordinal loss to learn the ordinal nature of AD viaeffectively integrating instance-to-instance ordinality, instance-to-classcompactness, and class-to-class separation. Third, to make the prototypelearning more stable, HOPE employs an exponential moving average strategy tolearn the global prototypes of NC and AD dynamically. Experimental results onthe internal ADNI and the external NACC datasets demonstrate the superiority ofthe proposed HOPE over existing state-of-the-art methods as well as itsinterpretability. Source code is made available athttps://github.com/thibault-wch/HOPE-for-mild-cognitive-impairment.

Alzheimer's disease (AD) is one of the most common neurodegenerative disorders presenting irreversible progression of cognitive impairment. How to identify AD as early as possible is critical for intervention with potential preventive measures. Among various neuroimaging modalities used to diagnose AD, functional positron emission tomography (PET) has higher sensitivity than structural magnetic resonance imaging (MRI), but it is also costlier and often not available in many hospitals. How to leverage massive unpaired unlabeled PET to improve the diagnosis performance of AD from MRI becomes rather important. To address this challenge, this paper proposes a novel joint learning framework of unsupervised cross-modal synthesis and AD diagnosis by mining underlying shared modality information, improving the AD diagnosis from MRI while synthesizing more discriminative PET images. We mine underlying shared modality information in two aspects: diversifying modality information through the cross-modal synthesis network and locating critical diagnosis-related patterns through the AD diagnosis network. First, to diversify the modality information, we propose a novel unsupervised cross-modal synthesis network, which implements the inter-conversion between 3D PET and MRI in a single model modulated by the AdaIN module. Second, to locate shared critical diagnosis-related patterns, we propose an interpretable diagnosis network based on fully 2D convolutions, which takes either 3D synthesized PET or original MRI as input. Extensive experimental results on the ADNI dataset show that our framework can synthesize more realistic images, outperform the state-of-the-art AD diagnosis methods, and have better generalization on external AIBL and NACC datasets.

Despite their impressive generative performance, latent diffusion model-based virtual try-on (VTON) methods lack faithfulness to crucial details of the clothes, such as style, pattern, and text. To alleviate these issues caused by the diffusion stochastic nature and latent supervision, we propose a novel Faithful Latent Diffusion Model for VTON, termed FLDM-VTON. FLDM-VTON improves the conventional latent diffusion process in three major aspects. First, we propose incorporating warped clothes as both the starting point and local condition, supplying the model with faithful clothes priors. Second, we introduce a novel clothes flattening network to constrain generated try-on images, providing clothes-consistent faithful supervision. Third, we devise a clothes-posterior sampling for faithful inference, further enhancing the model performance over conventional clothes-agnostic Gaussian sampling. Extensive experimental results on the benchmark VITON-HD and Dress Code datasets demonstrate that our FLDM-VTON outperforms state-of-the-art baselines and is able to generate photo-realistic try-on images with faithful clothing details.

Despite their impressive generative performance, latent diffusion model-based virtual try-on (VTON) methods lack faithfulness to crucial details of the clothes, such as style, pattern, and text. To alleviate these issues caused by the diffusion stochastic nature and latent supervision, we propose a novel Faithful Latent Diffusion Model for VTON, termed FLDM-VTON. FLDM-VTON improves the conventional latent diffusion process in three major aspects. First, we propose incorporating warped clothes as both the starting point and local condition, supplying the model with faithful clothes priors. Second, we introduce a novel clothes flattening network to constrain generated try-on images, providing clothes-consistent faithful supervision. Third, we devise a clothes-posterior sampling for faithful inference, further enhancing the model performance over conventional clothes-agnostic Gaussian sampling. Extensive experimental results on the benchmark VITON-HD and Dress Code datasets demonstrate that our FLDM-VTON outperforms state-of-the-art baselines and is able to generate photo-realistic try-on images with faithful clothing details.

Medical image segmentation is a challenging task with inherent ambiguity and high uncertainty attributed to factors such as unclear tumor boundaries and multiple plausible annotations. The accuracy and diversity of segmentation masks are both crucial for providing valuable references to radiologists in clinical practice. While existing diffusion models have shown strong capacities in various visual generation tasks, it is still challenging to deal with discrete masks in segmentation. To achieve accurate and diverse medical image segmentation masks, we propose a novel conditional Bernoulli Diffusion model for medical image segmentation (BerDiff). Instead of using the Gaussian noise, we first propose to use the Bernoulli noise as the diffusion kernel to enhance the capacity of the diffusion model for binary segmentation tasks, resulting in more accurate segmentation masks. Second, by leveraging the stochastic nature of the diffusion model, our BerDiff randomly samples the initial Bernoulli noise and intermediate latent variables multiple times to produce a range of diverse segmentation masks, which can highlight salient regions of interest that can serve as a valuable reference for radiologists. In addition, our BerDiff can efficiently sample sub-sequences from the overall trajectory of the reverse diffusion, thereby speeding up the segmentation process. Extensive experimental results on two medical image segmentation datasets with different modalities demonstrate that our BerDiff outperforms other recently published state-of-the-art methods. Source code is made available at https://github.com/takimailto/BerDiff .