In this paper, we introduce a trajectory planning problem for a marsupial robotics system consisting of a ground robot, a drone, and a taut tether of bounded length connecting the two robots. This problem can be framed within the context of a pursuit-evasion game. Using a geometric modeling approach, we present an optimal algorithm to compute a minimum-link path for the pursuer (ground robot), given the known path of the evader (drone). Furthermore, we address and solve three related geometric optimization problems, leveraging the intrinsic connections between them.
Accurate targeting is critical for the effectiveness of transcranial focused ultrasound (tFUS) neuromodulation. While CT provides accurate skull acoustic properties, its ionizing radiation and poor soft tissue contrast limit clinical applicability. In contrast, MRI offers superior neuroanatomical visualization without radiation exposure but lacks skull property mapping. This study proposes a novel, fully CT free simulation framework that integrates MRI-derived synthetic CT (sCT) with efficient modeling techniques for rapid and precise tFUS targeting. We trained a deep-learning model to generate sCT from T1-weighted MRI and integrated it with both full-wave (k-Wave) and accelerated simulation methods, hybrid angular spectrum (kWASM) and Rayleigh-Sommerfeld ASM (RSASM). Across five skull models, both full-wave and hybrid pipelines using sCT demonstrated sub-millimeter targeting deviation, focal shape consistency (FWHM ~3.3-3.8 mm), and <0.2 normalized pressure error compared to CT-based gold standard. Notably, the kW-ASM and RS-ASM pipelines reduced simulation time from ~3320 s to 187 s and 34 s respectively, achieving ~94% and ~90% time savings. These results confirm that MRI-derived sCT combined with innovative rapid simulation techniques enables fast, accurate, and radiation-free tFUS planning, supporting its feasibility for scalable clinical applications.
IceCube-Gen2 is a planned next-generation neutrino observatory at the South Pole that builds upon the successful design of IceCube. Integrating two complementary detection technologies for neutrinos, optical and radio Cherenkov emission, in combination with a surface array for cosmic-ray air shower detection, IceCube-Gen2 will cover a broad neutrino energy range from MeV to EeV. This index of contributions to the 39th International Cosmic Ray Conference in Geneva, Switzerland (July 15-24, 2025) describes research and development efforts for IceCube-Gen2. Included are summaries of the design, status, and sensitivity of the IceCube-Gen2 optical, surface, and radio components; performance studies of next-generation surface detectors and in-ice optical sensors; advanced reconstruction techniques of cosmic-ray air showers and neutrino events; sustainability and environmental impact; and sensitivity studies of astrophysical neutrino fluxes and cosmic-ray physics. Contributions related to IceCube and the scheduled IceCube Upgrade are available in a separate collection.
The IceCube Observatory at the South Pole has been operating in its full configuration since May 2011 with a duty cycle of about 99%. Its main component consists of a cubic-kilometer array of optical sensors deployed deep in the Glacial ice designed for the detection of high-energy astrophysical neutrinos. A surface array for cosmic ray air shower detection, IceTop, and a denser inner subdetector, DeepCore, significantly enhance the capabilities of the observatory, making it a multipurpose facility. This list of contributions to the 39th International Cosmic Ray Conference in Geneva, Switzerland (July 15-24, 2025) summarizes the latest results from IceCube covering a broad set of key questions in physics and astrophysics. The papers in this index are grouped topically to highlight IceCube contributions related to high-energy neutrino and multi-messenger astrophysics, atmospheric fluxes, cosmic-ray physics, low-energy neutrino transients, physics beyond the Standard Model, detector calibration and event reconstruction, and the status and performance of the IceCube Upgrade, a dense sensor infill complemented by calibration devices to be deployed by the end of 2025. Contributions related to IceCube-Gen2, the planned future extension of IceCube, are available in a separate collection.
Building on the current program's success and driven by new physics challenges, the NA61/SHINE Collaboration proposes to continue measuring hadron production properties in reactions induced by hadron and ion beams after CERN Long Shutdown 3. These measurements are of significant interest to the heavy-ion, cosmic-ray, and neutrino physics communities and will focus on: - Investigating hadron production in the light-ion systems to explore the diagram of high-energy nuclear collisions, and to obtain new insight into the unexpected violation of isospin (flavor) symmetry recently observed by the experiment; - Measuring charm-anticharm correlations to gain unique insights into the production locality of charm and anticharm quark pairs; - Examining strangeness and multi-strangeness production to improve our understanding of the early Universe's evolution and neutron star formation; - Measuring cross sections relevant for cosmic-ray measurements, significantly boosting searches for new physics in our Galaxy; - Conducting hadron production measurements with proton, pion, and kaon beams for neutrino physics, enhancing the precision of hadron production data needed for initial neutrino flux predictions in neutrino oscillation experiments; - Measuring hadron production processes relevant for understanding the flux of atmospheric neutrinos, as well as neutrinos and muons from spallation sources. To achieve these objectives, a detector upgrade and a beam upgrade are required, with data-taking planned for the period 2029-2032 and beyond.
Medical image retrieval is a valuable field for supporting clinical decision-making, yet current methods primarily support 2D images and require fully annotated queries, limiting clinical flexibility. To address this, we propose RadiomicsRetrieval, a 3D content-based retrieval framework bridging handcrafted radiomics descriptors with deep learning-based embeddings at the tumor level. Unlike existing 2D approaches, RadiomicsRetrieval fully exploits volumetric data to leverage richer spatial context in medical images. We employ a promptable segmentation model (e.g., SAM) to derive tumor-specific image embeddings, which are aligned with radiomics features extracted from the same tumor via contrastive learning. These representations are further enriched by anatomical positional embedding (APE). As a result, RadiomicsRetrieval enables flexible querying based on shape, location, or partial feature sets. Extensive experiments on both lung CT and brain MRI public datasets demonstrate that radiomics features significantly enhance retrieval specificity, while APE provides global anatomical context essential for location-based searches. Notably, our framework requires only minimal user prompts (e.g., a single point), minimizing segmentation overhead and supporting diverse clinical scenarios. The capability to query using either image embeddings or selected radiomics attributes highlights its adaptability, potentially benefiting diagnosis, treatment planning, and research on large-scale medical imaging repositories. Our code is available at https://github.com/nainye/RadiomicsRetrieval.
Identifying what aspects of brain activity are responsible for conscious perception remains one of the most challenging problems in science. While progress has been made through psychophysical studies employing EEG and fMRI, research would greatly benefit from improved methods for stimulating the brain in healthy human subjects. Traditional techniques for neural stimulation through the skull, including electrical or magnetic stimulation, suffer from coarse spatial resolution and have limited ability to target deep brain structures with high spatial selectivity. Over the past decade, a new tool has emerged known as transcranial focused ultrasound (tFUS), which enables the human brain to be stimulated safely and non-invasively through the skull with millimeter-scale spatial resolution, including cortical as well as deep brain structures. This tool offers an exciting opportunity for breakthroughs in consciousness research. Given the extensive preparation and regulatory approvals associated with tFUS testing, careful experimental planning is essential. Therefore, our goal here is to provide a roadmap for using tFUS in humans for exploring the neural substrate of conscious perception.
Since 2021, the Open Data Portal has provided access to the Pierre Auger Observatory's data for both the scientific community and the general public. The data release process has been in place since the Observatory's foundation. It continues to be strengthened as outlined in the approved policy and the Observatory's Data Management Plan. More than 80 000 cosmic-ray events above $10^{17}$ eV, detected with the surface and fluorescence detectors, have been released at various levels, from calibrated traces to high-level reconstruction parameters. Additionally, atmospheric data and low-energy particle counting rates have been made available for space weather studies. The Collaboration is committed to releasing FAIR (Findable, Accessible, Interoperable, and Reusable) data, along with accompanying software and detailed documentation, enabling users to perform their own queries and analyses for both research and educational purposes. These datasets have already served as a basis for several scientific papers and have been widely used in various outreach activities. After 20 years of stable data acquisition, the Pierre Auger Collaboration will disclose 30% of the cosmic ray events above $2.5 \cdot 10^{18}$ eV collected with the main surface detector array between 2004 and 2022, corresponding to an exposure of about 24 000 km$^2$ sr yr, together with events detected with the fluorescence detector and used for energy calibration. This release will provide an unprecedented public dataset for ultra-high-energy cosmic rays, enabling in-depth studies of their properties. Together with the published catalog of the 100 most energetic events recorded, this initiative represents the Pierre Auger Collaboration's strong commitment to distributed and collective knowledge, sharing progress with the entire scientific community.
A precise understanding of the optical properties of the instrumented Antarctic ice sheet is crucial to the performance of optical Cherenkov telescopes such as the IceCube Neutrino Observatory and its planned successor, IceCube-Gen2. One complication arising from the large envisioned footprint of IceCube-Gen2 is the larger impact of the so-called ice tilt. It describes the undulation of ice layers of constant optical properties within the detector. In this contribution, we will describe the project to build a co-deployed laser dust logger. This is a device to measure the stratigraphy of impurities in the ice to derive the ice tilt. It consists of a light source that will be co-deployed with the photosensor modules, meaning it is part of the deployment string and operated during the deployment of the detector. The newly developed device will be tested during the deployment of the IceCube Upgrade in the 2025/26 austral summer to pave the way for IceCube-Gen2.
LiDAR-based 3D mapping suffers from cumulative drift causing global misalignment, particularly in GNSS-constrained environments. To address this, we propose a unified framework that fuses LiDAR, GNSS, and IMU data for high-resolution city-scale mapping. The method performs velocity-based temporal alignment using Dynamic Time Warping and refines GNSS and IMU signals via extended Kalman filtering. Local maps are built using Normal Distributions Transform-based registration and pose graph optimization with loop closure detection, while global consistency is enforced using GNSS-constrained anchors followed by fine registration of overlapping segments. We also introduce a large-scale multimodal dataset captured in Perth, Western Australia to facilitate future research in this direction. Our dataset comprises 144{,}000 frames acquired with a 128-channel Ouster LiDAR, synchronized RTK-GNSS trajectories, and MEMS-IMU measurements across 21 urban loops. To assess geometric consistency, we evaluated our method using alignment metrics based on road centerlines and intersections to capture both global and local accuracy. Our method reduces the average global alignment error from 3.32\,m to 1.24\,m, achieving a 61.4\% improvement. The constructed high-fidelity map supports a wide range of applications, including smart city planning, geospatial data integration, infrastructure monitoring, and GPS-free navigation. Our method, and dataset together establish a new benchmark for evaluating 3D city mapping in GNSS-constrained environments. The dataset and code will be released publicly.
An innovative optical module (OM) with segmented light-sensitive area has been developed for IceCube-Gen2 that will take neutrino astronomy at the South Pole to the next level. It builds on the successful features of the mDOM and D-Egg modules of IceCube Upgrade while adapting to the smaller borehole diameter of IceCube-Gen2. The newly developed OM, which is being tested in IceCube Upgrade, serves as a prototype for the planned mass production of about 10,000 OMs for IceCube-Gen2. To simplify the assembly process, important changes were made to the design, in particular to integrate the new gel pad concept. This replaces the 3D-printed support structure of the mDOM while maintaining through total internal reflection the increased light collection efficiency of the reflector rings. In addition, the design features local generation of high voltage for each photomultiplier tube (PMT) via a Cockcroft-Walton circuit and the full digitization of the signal on each PMT base with a sampling rate of 60 MSpS. This significantly reduces the complexity of the mainboard so that it fits into the limited space available. This article describes the development status and presents the performance of the first prototypes.
Lane changes of preceding vehicles have a great impact on the motion planning of automated vehicles especially in complex traffic situations. Predicting them would benefit the public in terms of safety and efficiency. While many research efforts have been made in this direction, few concentrated on predicting maneuvers within a set time interval compared to predicting at a set prediction time. In addition, there exist a lack of comparisons between different architectures to try to determine the best performing one and to assess how to correctly choose the input for such models. In this paper the structure of an LSTM, a CNN and a Transformer network are described and implemented to predict the intention of human drivers to perform a lane change. We show how the data was prepared starting from a publicly available dataset (highD), which features were used, how the networks were designed and finally we compare the results of the three networks with different configurations of input data. We found that transformer networks performed better than the other networks and was less affected by overfitting. The accuracy of the method spanned from $82.79\%$ to $96.73\%$ for different input configurations and showed overall good performances considering also precision and recall.
Efficient planning, resource management, and consistent operations often rely on converting textual process documents into formal Business Process Model and Notation (BPMN) models. However, this conversion process remains time-intensive and costly. Existing approaches, whether rule-based or machine-learning-based, still struggle with writing styles and often fail to identify parallel structures in process descriptions. This paper introduces an automated pipeline for extracting BPMN models from text, leveraging the use of machine learning and large language models. A key contribution of this work is the introduction of a newly annotated dataset, which significantly enhances the training process. Specifically, we augment the PET dataset with 15 newly annotated documents containing 32 parallel gateways for model training, a critical feature often overlooked in existing datasets. This addition enables models to better capture parallel structures, a common but complex aspect of process descriptions. The proposed approach demonstrates adequate performance in terms of reconstruction accuracy, offering a promising foundation for organizations to accelerate BPMN model creation.
Accessibility, defined as travel impedance between spatially dispersed opportunities for activity, is one of the main determinants of public transport (PT) use. In-depth understanding of its properties is crucial for optimal public transport systems planning and design. Although the concept has been around for decades and there is a large body of literature on accessibility operationalisation and measurement, a unified approach is lacking. To this end, we introduce a novel graph representation of public transport networks, termed the access graph, based on the shortest paths between nodes. Shortest paths are calculated using the in-vehicle time-weighted L- and frequency-weighted P-space representations to determine generalised travel times. Then there is an edge between two nodes in the access graph if the travel time between them is below a certain threshold time budget. In this representation, node degree directly measures the number of nodes reachable within a predetermined time. We study the threshold-dependent evolution of the access graph, focusing on average degree and degree distributions. Based on the topological properties of the access graph, we define a set of accessibility indicators. In addition, we propose indicators of access equity. We apply the methodology to a dataset of 51 metro networks worldwide. In all cases, a logistic-like growth of average degree with time budget is observed, indicating universal behaviour of accessibility and exhibiting the value of the proposed representation for unified accessibility studies and its potential for comparative analyses. We see a great potential for the access graph to drive in-depth studies of accessibility.
The Patent-Based Idea Generation task asks systems to turn real patents into product ideas viable within three years. We propose MK2, a prompt-centric pipeline: Gemini 2.5 drafts and iteratively edits a prompt, grafting useful fragments from weaker outputs; GPT-4.1 then uses this prompt to create one idea per patent, and an Elo loop judged by Qwen3-8B selects the best prompt-all without extra training data. Across three domains, two evaluator types, and six criteria, MK2 topped the automatic leaderboard and won 25 of 36 tests. Only the materials-chemistry track lagged, indicating the need for deeper domain grounding; yet, the results show that lightweight prompt engineering has already delivered competitive, commercially relevant ideation from patents.
This paper addresses the significant challenge of recoil momentum accumulation during non-propulsive payload deployment for low-Earth-orbit mega-constellations. An efficient phase-based approximation algorithm is introduced, leading to over 90% faster computation while maintaining maximum ejection velocity errors below 1%. The analysis highlights that cumulative recoil velocity is the primary factor governing deployment capability, with excess velocity being approximately half of the cumulative recoil velocity. An analytical expression for predicting the maximum ejection velocity is derived, enabling rapid mission planning with less than 2% error. This framework establishes crucial operational boundaries concerning altitude differences, mass ratios, and deployment quantities for near-Earth orbits (up to 2,000 km), providing essential tools for the sustainable servicing of future constellations comprising over 200 satellites per orbital plane.
While total intracranial carotid artery calcification (ICAC) volume is an established stroke biomarker, growing evidence shows this aggregate metric ignores the critical influence of plaque location, since calcification in different segments carries distinct prognostic and procedural risks. However, a finer-grained, segment-specific quantification has remained technically infeasible. Conventional 3D models are forced to process downsampled volumes or isolated patches, sacrificing the global context required to resolve anatomical ambiguity and render reliable landmark localization. To overcome this, we reformulate the 3D challenge as a \textbf{Parallel Probabilistic Landmark Localization} task along the 1D axial dimension. We propose the \textbf{Depth-Sequence Transformer (DST)}, a framework that processes full-resolution CT volumes as sequences of 2D slices, learning to predict $N=6$ independent probability distributions that pinpoint key anatomical landmarks. Our DST framework demonstrates exceptional accuracy and robustness. Evaluated on a 100-patient clinical cohort with rigorous 5-fold cross-validation, it achieves a Mean Absolute Error (MAE) of \textbf{0.1 slices}, with \textbf{96\%} of predictions falling within a $\pm1$ slice tolerance. Furthermore, to validate its architectural power, the DST backbone establishes the best result on the public Clean-CC-CCII classification benchmark under an end-to-end evaluation protocol. Our work delivers the first practical tool for automated segment-specific ICAC analysis. The proposed framework provides a foundation for further studies on the role of location-specific biomarkers in diagnosis, prognosis, and procedural planning. Our code will be made publicly available.
The lack of high-quality public cyber incident data limits empirical research and predictive modeling for cyber risk assessment. This challenge persists due to the reluctance of companies to disclose incidents that could damage their reputation or investor confidence. Therefore, from an actuarial perspective, potential resolutions conclude two aspects: the enhancement of existing cyber incident datasets and the implementation of advanced modeling techniques to optimize the use of the available data. A review of existing data-driven methods highlights a significant lack of entity-specific organizational features in publicly available datasets. To address this gap, we propose a novel InsurTech framework that enriches cyber incident data with entity-specific attributes. We develop various machine learning (ML) models: a multilabel classification model to predict the occurrence of cyber incident types (e.g., Privacy Violation, Data Breach, Fraud and Extortion, IT Error, and Others) and a multioutput regression model to estimate their annual frequencies. While classifier and regressor chains are implemented to explore dependencies among cyber incident types as well, no significant correlations are observed in our datasets. Besides, we apply multiple interpretable ML techniques to identify and cross-validate potential risk factors developed by InsurTech across ML models. We find that InsurTech empowered features enhance prediction occurrence and frequency estimation robustness compared to only using conventional risk factors. The framework generates transparent, entity-specific cyber risk profiles, supporting customized underwriting and proactive cyber risk mitigation. It provides insurers and organizations with data-driven insights to support decision-making and compliance planning.
This paper presents a proposed AI Deep Learning model that addresses common challenges encountered in Visible Light Communication (VLC) systems. In this work, we run a Python simulation that models a basic VLC system primarily affected by Additive White Gaussian Noise (AWGN). A Deep Neural Network (DNN) is then trained to equalize the noisy signal received and improve signal integrity. The system evaluates and compares the Bit Error Rate (BER) before and after equalization to demonstrate the effectiveness of the proposed model. This paper starts by introducing the concept of visible light communication, then it dives deep into some details about the process of VLC and the challenges it faces, shortly after we propose our project which helps overcome these challenges. We finally conclude with a lead for future work, highlighting the areas that are most suitable for future improvements.
Unbalanced optimal transport (UOT) is a natural extension of optimal transport (OT) allowing comparison between measures of different masses. It arises naturally in machine learning by offering a robustness against outliers. The aim of this work is to provide convergence rates of the regularized transport cost and plans towards their original solution when both measures are weighted sums of Dirac masses.
In this paper, we address a bicriteria network design problem that arises from practical applications in urban and rural public transportation planning. We establish the problem's complexity and demonstrate inapproximability results, highlighting the inherent difficulties in finding optimal solutions. Additionally, we identify special cases where approximability can be achieved, providing valuable insights for practitioners. Our proofs leverage complexity results related to directed network design problems, an area that has received limited attention in the existing literature. By investigating these complexity results, we aim to fill a critical gap and enhance the understanding of the interplay between bicriteria decision-making and network design challenges.
Neural networks (NNs) have a great potential for future neutrino telescopes such as IceCube-Gen2, the planned high-energy extension of the IceCube observatory. IceCube-Gen2 will feature new optical sensors with multiple photomultiplier tubes (PMTs) designed to provide omnidirectional sensitivity. Neural networks excel at handling high-dimensional problems and can naturally incorporate the increased complexity of these new sensors. Additionally, their fast inference time makes them promising candidates for handling the high event rates expected from IceCube-Gen2. This contribution presents potential applications of neural networks in the IceCube-Gen2 in-ice optical array. First, we introduce a method to simulate the IceCube-Gen2 optical modules' photon acceptance using a NN that leverages the modules' inherent symmetries. Secondly, we present the status of neutrino NN-based reconstruction efforts, including the adaptation of a novel IceCube technique that combines normalizing flows with transformer NNs. Finally, we describe current progress in noise cleaning applications based on node classification with graph neural networks (GNNs), a method that has already shown promising results for the forthcoming low-energy extension, IceCube-Upgrade.
Ensuring reliable and low-latency communication in offshore wind farms is critical for efficient monitoring and control, yet remains challenging due to the harsh environment and lack of infrastructure. This paper investigates a flying base station (FBS) approach for wide-area monitoring and control in the UK Hornsea offshore wind farm project. By leveraging mobile, flexible FBS platforms in the remote and harsh offshore environment, the proposed system offers real-time connectivity for turbines without the need for deploying permanent infrastructure at the sea. We develop a detailed and practical end-to-end latency model accounting for five key factors: flight duration, connection establishment, turbine state information upload, computational delay, and control transmission, to provide a holistic perspective often missing in prior studies. Furthermore, we combine trajectory planning, beamforming, and resource allocation into a multi-objective optimization framework for the overall latency minimization, specifically designed for large-scale offshore wind farm deployments. Simulation results verify the effectiveness of our proposed method in minimizing latency and enhancing efficiency in FBS-assisted offshore monitoring across various power levels, while consistently outperforming baseline designs.
The next generation neutrino telescope, IceCube-Gen2, will be sensitive to the astrophysical and cosmogenic flux of neutrinos across a broad energy range, from the TeV to the EeV scale. The planned design includes 8 cubic kilometers of ice instrumented with approximately 10,000 optical sensors, a surface array, and a radio array of antennas embedded in the ice laid out sparsely over 500 km^2. The radio array provides sensitivity to ultra-high energy neutrinos using independent radio stations that can trigger on Askaryan emission from neutrino interactions in the ice. In this contribution, we present the design for the radio array along with its planned implementation, which is expected to increase sensitivity to neutrinos with energies beyond 100PeV by at least an order of magnitude over existing arrays. Furthermore, we will quantify the expected science output by presenting measurement forecasts for the main science cases of diffuse flux and point source discovery, as well as cross-section and flavor measurements.
Background: Accurate forecasting of lung tumor motion is essential for precise dose delivery in proton therapy. While current markerless methods mostly rely on deep learning, transformer-based architectures remain unexplored in this domain, despite their proven performance in trajectory forecasting. Purpose: This work introduces a markerless forecasting approach for lung tumor motion using Vision Transformers (ViT). Two training strategies are evaluated under clinically realistic constraints: a patient-specific (PS) approach that learns individualized motion patterns, and a multi-patient (MP) model designed for generalization. The comparison explicitly accounts for the limited number of images that can be generated between planning and treatment sessions. Methods: Digitally reconstructed radiographs (DRRs) derived from planning 4DCT scans of 31 patients were used to train the MP model; a 32nd patient was held out for evaluation. PS models were trained using only the target patient's planning data. Both models used 16 DRRs per input and predicted tumor motion over a 1-second horizon. Performance was assessed using Average Displacement Error (ADE) and Final Displacement Error (FDE), on both planning (T1) and treatment (T2) data. Results: On T1 data, PS models outperformed MP models across all training set sizes, especially with larger datasets (up to 25,000 DRRs, p < 0.05). However, MP models demonstrated stronger robustness to inter-fractional anatomical variability and achieved comparable performance on T2 data without retraining. Conclusions: This is the first study to apply ViT architectures to markerless tumor motion forecasting. While PS models achieve higher precision, MP models offer robust out-of-the-box performance, well-suited for time-constrained clinical settings.
Digital Twin (DT) technology has emerged as a transformative approach in healthcare, but its application in personalized patient care remains limited. This paper aims to present a practical implementation of DT in the management of chronic diseases. We introduce a general DT framework for personalized care planning (DT4PCP), with the core components being a real-time virtual representation of a patient's health and emerging predictive models to enable adaptive, personalized care. We implemented the DT4PCP framework for managing Type 2 Diabetes (DT4PCP-T2D), enabling real-time collection of behavioral data from patients with T2D, predicting emergency department (ED) risks, simulating the effects of different interventions, and personalizing care strategies to reduce ED visits. The DT4PCP-T2D also integrates social determinants of health (SDoH) and other contextual data, offering a comprehensive view of the patient's health to ensure that care recommendations are tailored to individual needs. Through retrospective simulations, we demonstrate that integrating DTs in T2D management can lead to significant advancements in personalized medicine. This study underscores the potential of DT technology to revolutionize chronic disease care.
Accurate survival prediction is critical in oncology for prognosis and treatment planning. Traditional approaches often rely on a single data modality, limiting their ability to capture the complexity of tumor biology. To address this challenge, we introduce a multimodal deep learning framework for survival analysis capable of modeling both single and competing risks scenarios, evaluating the impact of integrating multiple medical data sources on survival predictions. We propose SAMVAE (Survival Analysis Multimodal Variational Autoencoder), a novel deep learning architecture designed for survival prediction that integrates six data modalities: clinical variables, four molecular profiles, and histopathological images. SAMVAE leverages modality specific encoders to project inputs into a shared latent space, enabling robust survival prediction while preserving modality specific information. Its parametric formulation enables the derivation of clinically meaningful statistics from the output distributions, providing patient-specific insights through interactive multimedia that contribute to more informed clinical decision-making and establish a foundation for interpretable, data-driven survival analysis in oncology. We evaluate SAMVAE on two cancer cohorts breast cancer and lower grade glioma applying tailored preprocessing, dimensionality reduction, and hyperparameter optimization. The results demonstrate the successful integration of multimodal data for both standard survival analysis and competing risks scenarios across different datasets. Our model achieves competitive performance compared to state-of-the-art multimodal survival models. Notably, this is the first parametric multimodal deep learning architecture to incorporate competing risks while modeling continuous time to a specific event, using both tabular and image data.
Mandibular Angle Split Osteotomy (MASO) is a significant procedure in oral and maxillofacial surgery. Despite advances in technique and instrumentation, its success still relies heavily on the surgeon's experience. In this work, a human-robot collaborative system is proposed to perform MASO according to a preoperative plan and under guidance of a surgeon. A task decomposition methodology is used to divide the collaborative surgical procedure into three subtasks: (1) positional control and (2) orientation control, both led by the robot for precise alignment; and (3) force-control, managed by surgeon to ensure safety. Additionally, to achieve patient tracking without the need for a skull clamp, an optical tracking system (OTS) is utilized. Movement of the patient mandibular is measured with an optical-based tracker mounted on a dental occlusal splint. A registration method and Robot-OTS calibration method are introduced to achieve reliable navigation within our framework. The experiments of drilling were conducted on the realistic phantom model, which demonstrated that the average error between the planned and actual drilling points is 1.85mm.
Understanding and predicting mobility dynamics in transportation networks is critical for infrastructure planning, resilience analysis, and traffic management. Traditional graph-based models typically assume memoryless movement, limiting their ability to capture sequential dependencies inherent in real-world mobility patterns. In this study, we introduce a novel higher-order network framework for modeling memory-dependent dynamics in transportation systems. By extending classical graph representations through higher-order Markov chains and de Bruijn graph structures, our framework encodes the spatial and temporal ordering of traversed paths, enabling the analysis of structurally and functionally critical components with improved fidelity. We generalize key network analytics, including betweenness centrality, PageRank, and next-step prediction, to this higher-order setting and validate our approach on the Sioux Falls transportation network using agent-based trajectory data generated with MATSim. Experimental results demonstrate that higher-order models outperform first-order baselines across multiple tasks, with the third-order model achieving an optimal balance between predictive accuracy and model complexity. These findings highlight the importance of incorporating memory effects into network-based transportation analysis and offer a scalable, data-driven methodology for capturing complex mobility behaviors in infrastructure systems.
Medical Image Segmentation (MIS) plays a crucial role in medical therapy planning and robot navigation. Prototype learning methods in MIS focus on generating segmentation masks through pixel-to-prototype comparison. However, current approaches often overlook sample diversity by using a fixed prototype per semantic class and neglect intra-class variation within each input. In this paper, we propose to generate instance-adaptive prototypes for MIS, which integrates a common prototype proposal (CPP) capturing common visual patterns and an instance-specific prototype proposal (IPP) tailored to each input. To further account for the intra-class variation, we propose to guide the IPP generation by re-weighting the intermediate feature map according to their confidence scores. These confidence scores are hierarchically generated using a transformer decoder. Additionally we introduce a novel self-supervised filtering strategy to prioritize the foreground pixels during the training of the transformer decoder. Extensive experiments demonstrate favorable performance of our method.