For the past decades medical robotic solutions were mostly based on the concept of tele-manipulation. While their design was extremely intelligent, allowing for better access, improved dexterity, reduced tremor, and improved imaging, their intelligence was limited. They therefore left cognition and decision making to the surgeon. As medical robotics advances towards high-level autonomy, the scientific community needs to explore the required pathway towards partial and full autonomy. Here, we introduce the concept of Dyadic Partnership(DP), a new paradigm in which robots and clinicians engage in intelligent, expert interaction and collaboration. The Dyadic Partners would discuss and agree on decisions and actions during their dynamic and interactive collaboration relying also on intuitive advanced media using generative AI, such as a world model, and advanced multi-modal visualization. This article outlines the foundational components needed to enable such systems, including foundation models for clinical intelligence, multi-modal intent recognition, co-learning frameworks, advanced visualization, and explainable, trust-aware interaction. We further discuss key challenges such as data scarcity, lack of standardization, and ethical acceptance. Dyadic partnership is introduced and is positioned as a powerful yet achievable, acceptable milestone offering a promising pathway toward safer, more intuitive collaboration and a gradual transition to full autonomy across diverse clinical settings.

Visual effects (VFX) are key to immersion in modern films, games, and AR/VR. Creating 3D effects requires specialized expertise and training in 3D animation software and can be time consuming. Generative solutions typically rely on computationally intense methods such as diffusion models which can be slow at 4D inference. We reformulate 3D animation as a field prediction task and introduce a text-driven framework that infers a time-varying 4D flow field acting on 3D Gaussians. By leveraging large language models (LLMs) and vision-language models (VLMs) for function generation, our approach interprets arbitrary prompts (e.g., “make the vase glow orange, then explode”) and instantly updates color, opacity, and positions of 3D Gaussians in real time. This design avoids overheads such as mesh extraction, manual or physics-based simulations and allows both novice and expert users to animate volumetric scenes with minimal effort on a consumer device even in a web browser11Web Demo: http://promptvfx.duckdns.org. Experimental results show that simple textual instructions suffice to generate compelling time-varying VFX, reducing the manual effort typically required for rigging or advanced modeling. We thus present a fast and accessible pathway to language-driven 3D content creation that can pave the way to democratize VFX further. Code available at https://obsphera.github.io/promptvfx/.

Fine-grained understanding of operating room (OR) activity could enable workflow-aware assistance, yet remains difficult due to clutter, occlusions, and limited sensing. The prevailing approach to model this environment is scene graphs as an interpretable representation of OR interactions. Converting their frame-wise relational predictions into temporally extended, fine-grained actions however, is challenging without explicit temporal modeling. To enable a principled temporal evaluation of current OR understanding methods, we introduce the first action-centric benchmark built on a publicly available ego-exocentric OR dataset by defining a fine-grained, multi-role action taxonomy and generating dense action segments via distillation from ground-truth scene graph state changes. Experiments on this benchmark show that current scene graph prediction methods struggle to model temporal structure, even when adding explicit modeling through Graph Neural Networks. We therefore introduce a vision-only temporal model that outperforms graph-based methods significantly when using all available egocentric video as input. Building on this model we also introduce a novel multi- to single-view feature alignment strategy that improves single-view performance on multi-role action recognition, mitigating the need for extensive egocentric video capture. Benchmark and code will be released upon acceptance.

Spatiotemporal reasoning is a fundamental capability for artificial intelligence (AI) in soft tissue surgery, paving the way for intelligent assistive systems and autonomous robotics. While 2D vision-language models show increasing promise at understanding surgical video, the spatial complexity of surgical scenes suggests that reasoning systems may benefit from explicit 4D representations. Here, we propose a framework for equipping surgical agents with spatiotemporal tools based on an explicit 4D representation, enabling AI systems to ground their natural language reasoning in both time and 3D space. Leveraging models for point tracking, depth, and segmentation, we develop a coherent 4D model with spatiotemporally consistent tool and tissue semantics. A Multimodal Large Language Model (MLLM) then acts as an agent on tools derived from the explicit 4D representation (e.g., trajectories) without any fine-tuning. We evaluate our method on a new dataset of 134 clinically relevant questions and find that the combination of a general purpose reasoning backbone and our 4D representation significantly improves spatiotemporal understanding and allows for 4D grounding. We demonstrate that spatiotemporal intelligence can be "assembled" from 2D MLLMs and 3D computer vision models without additional training. Code, data, and examples are available at https://tum-ai.github.io/surg4d/

We propose a geometric framework for longitudinal multi-parametric MRI analysis based on patient-specific energy modelling in sequence space. Rather than operating on images with spatial networks, each voxel is represented by its multi-sequence intensity vector (T1, T1c, T2, FLAIR, ADC), and a compact implicit neural representation is trained via denoising score matching to learn an energy function E_θ(𝐮) over ℝ^d from a single baseline scan. The learned energy landscape provides a differential-geometric description of tissue regimes without segmentation labels. Local minima define tissue basins, gradient magnitude reflects proximity to regime boundaries, and Laplacian curvature characterises local constraint structure. Importantly, this baseline energy manifold is treated as a fixed geometric reference: it encodes the set of contrast combinations observed at diagnosis and is not retrained at follow-up. Longitudinal assessment is therefore formulated as evaluation of subsequent scans relative to this baseline geometry. Rather than comparing anatomical segmentations, we analyse how the distribution of MRI sequence vectors evolves under the baseline energy function. In a paediatric case with later recurrence, follow-up scans show progressive deviation in energy and directional displacement in sequence space toward the baseline tumour-associated regime before clear radiological reappearance. In a case with stable disease, voxel distributions remain confined to established low-energy basins without systematic drift. The presented cases serve as proof-of-concept that patient-specific energy manifolds can function as geometric reference systems for longitudinal mpMRI analysis without explicit segmentation or supervised classification, providing a foundation for further investigation of manifold-based tissue-at-risk tracking in neuro-oncology.

Percutaneous intervention procedures require the insertion and manipulation of a puncture instrument within the patient's body. In this context, a remote center-of-motion (RCM) parallel mechanism, termed PercuCob-RCM, is designed to support clinicians during needle-based interventions by securely holding the puncture instrument and enabling a translational motion along the direction of the needle and rotational motions about a fixed insertion point. A key challenge in developing this mechanism lies in the actuation of its universal joint. This work investigates three alternative actuation schemes and evaluates them to guide the construction of a suitable prototype. The prototype design is formulated as a biobjective optimization problem. The first objective function represents the overall size of the mechanism, while the second captures the motor torques required to sustain a specified axial tool force. For each actuation scheme, the set of nondominated Pareto-optimal solutions is computed and analyzed. Based on the comparative assessment of the Pareto fronts, a preferred design configuration is selected, and a functional prototype is developed.

Background Increasing evidence shows that intermuscular adipose tissue (IMAT) and lean muscle mass (LMM) influence cardiometabolic health; however, their independent and/or combined associations with cardiovascular risk in individuals without pre-existing conditions remain unclear. Purpose To assess whether IMAT and LMM are associated with cardiometabolic risk factors in individuals without pre-existing conditions. Materials and Methods A total of 11 348 participants (6460 [56.9%] men; median age, 43.0 years; IQR, 33.5-52.5 years) without any known pre-existing conditions underwent whole-body 3-T MRI as part of a prospective multicenter population study (German National Cohort, or NAKO). LMM and IMAT were quantified on MRI-based paraspinal muscle segmentations with a deep learning model. Cardiometabolic risk factors (hypertension, dysglycemia, and atherogenic dyslipidemia) were defined on the basis of laboratory test results and clinical examinations. Age- and sex-corrected z scores of LMM and IMAT were calculated. Associations of LMM and IMAT percentage with physical activity and cardiometabolic risk factors were examined with univariable and multivariable analyses. Results The percentage of IMAT increased with age and was greater in women, whereas LMM decreased with age and was lower in women. After adjustments for age, sex, and study site, increased IMAT was associated with increased odds of hypertension (odds ratio [OR], 1.67; 95% CI: 1.49, 1.86; P < .001), atherogenic dyslipidemia (OR, 1.82; 95% CI: 1.65, 2.00; P < .001), and dysglycemia (OR, 0.51; 95% CI: 0.35, 0.76; P = .009) in both sexes, whereas increased LMM was associated with decreased odds of all risk factors (dysglycemia: OR, 0.51; 95% CI: 0.35, 0.76; P = .009; atherogenic dyslipidemia: OR, 0.49; 95% CI: 0.39, 0.62; P < .001; hypertension: OR, 0.34; 95% CI: 0.24, 0.48; P < .001) in male participants only. Across z score combinations, participants with higher IMAT and lower LMM showed the highest prevalence of cardiometabolic risk factors. Conclusion IMAT and LMM, assessed on MRI scans, were independently associated with cardiometabolic risk factors in individuals without pre-existing conditions. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Hu in this issue. See also the editorial by Mohajer and Bari in this issue.

Foundation segmentation models excel at generating high-quality, class-agnostic masks, but they struggle to associate these proposals with specific target objects. This semantic gap severely hinders their deployment in downstream applications like robotic manipulation, which demand precise unseen objects segmentation. Existing approaches attempt to resolve this by relying on exhaustive 3D object model priors, inherently introducing prohibitive computational overhead and complex, multi-stage pipelines. To address these limitations, we propose SOS (Streamlined Object-conditional Transformer for model-free Segmentation). SOS completely eliminates the reliance on 3D models, requiring only a single reference image per target object. Central to our framework is a novel Object-Conditional Transformer that learns identity-anchored queries, unifying mask generation and target identification into a single feed-forward pass. This streamlined design drastically improves both structural and computational efficiency. Extensive evaluations across multiple benchmarks demonstrate that SOS establishes a new state-of-the-art for model-free unseen objects segmentation, delivering accurate and high-efficiency performance. The project page and code are available at https://sos-seg.github.io/.

Purpose: Detailed surgical recognition is critical for advancing AI-assisted surgery, yet progress is hampered by high annotation costs, data scarcity, and a lack of interpretable models. While scene graphs offer a structured abstraction of surgical events, their full potential remains untapped. In this work, we introduce ProtoFlow, a novel framework that learns dynamic scene graph prototypes to model complex surgical workflows in an interpretable and robust manner. Methods: ProtoFlow leverages a graph neural network (GNN) encoder-decoder architecture that combines self-supervised pretraining for rich representation learning with a prototype-based fine-tuning stage. This process discovers and refines core prototypes that encapsulate recurring, clinically meaningful patterns of surgical interaction, forming an explainable foundation for workflow analysis. Results: We evaluate our approach on the fine-grained CAT-SG dataset. ProtoFlow not only outperforms standard GNN baselines in overall accuracy but also demonstrates exceptional robustness in limited-data, few-shot scenarios, maintaining strong performance when trained on as few as one surgical video. Our qualitative analyses further show that the learned prototypes successfully identify distinct surgical sub-techniques and provide clear, interpretable insights into workflow deviations and rare complications. Conclusion: By uniting robust representation learning with inherent explainability, ProtoFlow represents a significant step toward developing more transparent, reliable, and data-efficient AI systems, accelerating their potential for clinical adoption in surgical training, real-time decision support, and workflow optimization.

Self-supervised depth estimation from monocular sequences relies on the joint learning of a depth and a pose network. Despite abundant research done to improve the depth network, efforts on the pose remain limited. In this context, even when depth is estimated up to scale, we highlight the importance of the alignment between the scene scales estimated by the pose and depth nets. Then, we introduce SA4Depth, an approach to improve this alignment and boost the depth predictions while keeping the inference time unchanged. Our proposed method uses the depth estimated during training to reproject learnable visual features across consecutive frames and refine the pose estimates by reducing feature alignment residuals. With our method, the estimated scene scales by the separate depth and pose networks are aligned, and the prediction scale consistency is improved across different sequences. Our differentiable refinement integrates seamlessly into existing self-supervised pipelines and substantially improves their depth estimates. We demonstrate this with extensive experiments both outdoors and indoors on KITTI, Cityscapes, and NYUv2. Additionally, results on KITTI Odometry confirm the effectiveness of our pose refinement.

Ultrasound-guided needle interventions are widely used in clinical practice, but their success critically depends on accurate needle placement, which is frequently hindered by the poor and intermittent visibility of needles in ultrasound images. Existing approaches remain limited by artifacts, occlusions, and low contrast, and often fail to support real-time continuous insertion. To overcome these challenges, this study introduces a robust real-time framework for continuous needle detection. We present ConVibNet, an extension of VibNet for detecting needles with significantly reduced visibility, addressing real-time, continuous needle tracking during insertion. ConVibNet leverages temporal dependencies across successive ultrasound frames to enable continuous estimation of both needle tip position and shaft angle in dynamic scenarios. To strengthen temporal awareness of needle tip motion, we introduce a novel intersection and difference loss that explicitly leverages motion correlations across consecutive frames. In addition, we curated a dedicated dataset for model development and evaluation. The performance of the proposed ConVibNet model was evaluated on our dataset, demonstrating superior accuracy compared to the baseline VibNet and UNet–LSTM models. Specifically, ConVibNet achieved a tip error of 2.80± 2.42 mm and an angle error of 1.69^∘± 2.00^∘ . These results represent a 0.75 mm improvement in tip localization accuracy over the best-performing baseline, while preserving real-time inference capability. ConVibNet advances real-time needle detection in ultrasound-guided interventions by integrating temporal correlation modeling with a novel intersection and difference loss, thereby improving accuracy and robustness and demonstrating high potential for integration into autonomous insertion systems.

Autonomous medical robots hold promise to improve patient outcomes, reduce provider workload, democratize access to care, and enable superhuman precision. However, autonomous medical robotics has been limited by a fundamental data problem: existing medical robotic datasets are small, single-embodiment, and rarely shared openly, restricting the development of foundation models that the field needs to advance. We introduce Open-H-Embodiment, the largest open dataset of medical robotic video with synchronized kinematics to date, spanning more than 49 institutions and multiple robotic platforms including the CMR Versius, Intuitive Surgical's da Vinci, da Vinci Research Kit (dVRK), Rob Surgical BiTrack, Virtual Incision's MIRA, Moon Surgical Maestro, and a variety of custom systems, spanning surgical manipulation, robotic ultrasound, and endoscopy procedures. We demonstrate the research enabled by this dataset through two foundation models. GR00T-H is the first open foundation vision-language-action model for medical robotics, which is the only evaluated model to achieve full end-to-end task completion on a structured suturing benchmark (25

The presence of adversarial examples in the physical world poses significant challenges to the deployment of Deep Neural Networks in safety-critical applications such as autonomous driving. Most existing methods for crafting physical-world adversarial examples are ad-hoc, relying on temporary modifications like shadows, laser beams, or stickers that are tailored to specific scenarios. In this paper, we introduce a new class of physical-world adversarial examples, AdvWT, which draws inspiration from the naturally occurring phenomenon of `wear and tear', an inherent property of physical objects. Unlike manually crafted perturbations, `wear and tear' emerges organically over time due to environmental degradation, as seen in the gradual deterioration of outdoor signboards. To achieve this, AdvWT follows a two-step approach. First, a GAN-based, unsupervised image-to-image translation network is employed to model these naturally occurring damages, particularly in the context of outdoor signboards. The translation network encodes the characteristics of damaged signs into a latent `damage style code'. In the second step, we introduce adversarial perturbations into the style code, strategically optimizing its transformation process. This manipulation subtly alters the damage style representation, guiding the network to generate adversarial images where the appearance of damages remains perceptually realistic, while simultaneously ensuring their effectiveness in misleading neural networks. Through comprehensive experiments on two traffic sign datasets, we show that AdvWT effectively misleads DNNs in both digital and physical domains. AdvWT achieves an effective attack success rate, greater robustness, and a more natural appearance compared to existing physical-world adversarial examples. Additionally, integrating AdvWT into training enhances a model's generalizability to real-world damaged signs.

Percutaneous epicardial access (PEA), performed on a beating heart under fluoroscopy, enables arrhythmia treatment. However, advancing a needle toward the thin, moving pericardium remains highly challenging and risky. To address this challenge, we present a novel physics-driven sonification method for Extended Reality (XR)-based multisensory navigation to enhance user perception regarding critically sensitive needle landing phase in PEA. Dynamic cardiac anatomy from 4DCTA was reconstructed and registered to a real-world coordinate system. Real-time needle tracking provided data of the needle tip relative to moving cardiac structures, driving the audio-visual feedback module. The visual display presents navigational cues and dynamic cardiac structures, whereas the auditory display encodes physiological cardiac states into a rigorous mapping through a multilayered physical membrane. To prove this novel concept, a phantom study was conducted with twelve cardiologists performing needle insertions under visual-only and multisensory feedback. Results show that the proposed multisensory method improved the navigation safety (chi(2) = 11.30, p < 0.01), reducing myocardial contact (3.64% vs. 7.27%) and increasing correct access (90.91% vs. 52.73%). Needle placement accuracy improved, with closer membrane proximity (Cliff's delta = 0.19) and reduced variability (Fligner p < 0.05). Execution time remained comparable, but time-accuracy correlations differed significantly between modalities (Fisher's p < 0.01). NASA-TLX showed lower cognitive load with multisensory guidance (t = -4.02, p < 0.01). In conclusion, the multisensory system is associated with improved spatiotemporal awareness and suggests potential benefits for surgical navigation, while indicating reduced cognitive load. These findings demonstrate the feasibility of physics-driven sonification for supporting user-centered interaction and motivate further validation in realistic and clinical environments.

The integration of multimodal imaging into operating rooms paves the way for comprehensive surgical scene understanding. In ophthalmic surgery, by now, two complementary imaging modalities are available: operating microscope (OPMI) imaging and real-time intraoperative optical coherence tomography (iOCT). This first work toward temporal OPMI and iOCT feature fusion demonstrates the potential of multimodal image processing for multi-head prediction through the example of precise instrument tracking in vitreoretinal surgery. We propose a multimodal, temporal, real-time capable network architecture to perform joint instrument detection, keypoint localization, and tool-tissue distance estimation. Our network design integrates a cross-attention fusion module to merge OPMI and iOCT image features, which are efficiently extracted via a Yolo-NAS and a CNN encoder, respectively. Furthermore, a region-based recurrent module leverages temporal coherence. Our experiments demonstrate reliable instrument localization and keypoint detection (95.79 284 m (OPMI only) to 33 m (multimodal). Feature fusion of multimodal imaging can enhance multitask prediction accuracy compared to single-modality processing, and real-time processing performance can be achieved through tailored network design. While our results demonstrate the potential of multimodal processing for image-guided vitreoretinal surgery, they also underline key challenges that motivate future research toward more reliable, consistent, and comprehensive surgical scene understanding.

Forecasting the evolution of dynamic environments is crucial for autonomous agents. While generative world models have recently achieved high photorealism in 2D video synthesis, by mixing within the image plane ego-motion and environmental dynamics, they exhibit physical inconsistencies, such as morphing or vanishing objects, especially over long time horizons. In this paper, we propose FR3D, a world model that predicts a persistent 3D latent representation for future dynamic 3D reconstruction. Unlike prior works that treat the world as a sequence of image-based features, FR3D explicitly decouples the 3D evolution of the scene from the agent's trajectory, treating the inferred ego-motion as a latent proxy for action. This disentanglement resolves the ambiguities between self-motion and world-motion, ensuring geometric consistency into the future. Furthermore, we introduce a teacher-student distillation strategy that leverages the spatial "common sense" of off-the-shelf foundation models, leading to robust zero-shot generalization. Extensive experiments demonstrate FR3D's strong performance for future dynamic 3D reconstruction from monocular observations across multiple datasets, even 2 seconds into the future.

Acute poly-substance intoxication requires rapid, life-saving decisions under substantial uncertainty, as clinicians must rely on incomplete ingestion details and nonspecific symptoms. Effective diagnostic reasoning in this chaotic environment requires fusing unstructured, non-medical narratives (e.g. paramedic scene descriptions and unreliable patient self-reports or known histories), with structured medical data like vital signs. While Large Language Models (LLMs) show potential for processing such heterogeneous inputs, they struggle in this setting, often underperforming simple baselines that rely solely on patient histories. To address this, we present DeToxR (Decision-support for Toxicology with Reasoning), the first adaptation of Reinforcement Learning (RL) to emergency toxicology. We design a robust data-fusion engine for multi-label prediction across 14 substance classes based on an LLM finetuned with Group Relative Policy Optimization (GRPO). We optimize the model's reasoning directly using a clinical performance reward. By formulating a multi-label agreement metric as the reward signal, the model is explicitly penalized for missing co-ingested substances and hallucinating absent poisons. Our model significantly outperforms its unadapted base LLM counterpart and supervised baselines. Furthermore, in a clinical validation study, the model indicates a clinical advantage by outperforming an expert toxicologist in identifying the correct poisons (Micro-F1: 0.644 vs. 0.473). These results demonstrate the potential of RL-aligned LLMs to synthesize unstructured pre-clinical narratives and structured medical data for decision support in high-stakes environments.

Vision-Language Models (VLMs) have significantly advanced medical visual question answering, yet their performance in ultrasound remains suboptimal. In clinical practice, sonographers explicitly focus on lesion regions to formulate reports, though diagnostic interpretations sometimes vary due to inherent subjectivity. However, existing VLMs are not explicitly structured to interactively zoom into lesions prior to diagnosis; moreover, they typically treat annotations as unbiased ground truths, failing to account for their inherent subjectivity and ambiguity. In this paper, we propose a framework specifically designed to consider the sonographer's cognitive workflow. We first introduce a structured Zoom-then-Diagnose paradigm, which replicates the interactive search process to enable lesion-focused reasoning. Furthermore, within the Group Relative Policy Optimization (GRPO) framework, we introduce an uncertainty-aware reward derived from stochastic group-wise rollouts to estimate prediction consistency as a proxy for model confidence. Together, these two components encourage the model to reinforce accurate predictions on clear cases while remaining cautious under ambiguity. Experiments across liver, breast, and thyroid datasets show that our framework improves lesion localization by 39.3%, demonstrating that our model has learned the ability to actively look closer and diagnose.

Few-shot anomaly detection streamlines and simplifies industrial safety inspection. However, limited samples make accurate differentiation between normal and abnormal features challenging, and even more so under category-agnostic conditions. Large-scale pre-training of foundation visual encoders has advanced many fields, as the enormous quantity of data helps to learn the general distribution of normal images. We observe that the anomaly amount in an image directly correlates with the difference in the learnt embeddings and utilize this to design a few-shot anomaly detector termed FoundAD. This is done by learning a nonlinear projection operator onto the natural image manifold. The simple operator acts as an effective tool for anomaly detection to characterize and identify out-of-distribution regions in an image. Extensive experiments show that our approach supports multi-class detection and achieves competitive performance compared to other approaches, while surpassing them in model size and inference efficiency. Backed up by evaluations with multiple foundation encoders, including fresh DINOv3, we believe this idea broadens the perspective on foundation features and advances the field of few-shot anomaly detection. Our code will be made public.

Estimating correspondences between deformed shape instances is a long-standing problem in computer graphics; numerous applications, from texture transfer to statistical modelling, rely on recovering an accurate correspondence map. Many methods have thus been proposed to tackle this challenging problem from varying perspectives, depending on the downstream application. This state-of-the-art report is geared towards researchers, practitioners, and students seeking to understand recent trends and advances in the field. We categorise developments into three paradigms: spectral methods based on functional maps, combinatorial formulations that impose discrete constraints, and deformation-based methods that directly recover a global alignment. Each school of thought offers different advantages and disadvantages, which we discuss throughout the report. Meanwhile, we highlight the latest developments in each area and suggest new potential research directions. Finally, we provide an overview of emerging challenges and opportunities in this growing field, including the recent use of vision foundation models for zero-shot correspondence and the particularly challenging task of matching partial shapes.