arXiv subjects

Linwei Wang

Publications and source records attributed to Linwei Wang.

At least 19 recordsLinked to original sources

arXiv2026

Single-pulse reanalysis of the 2024 Vela glitch and new observations of PSR~J0437$-$4715 and PSR~J1644$-$4559

The Pulsar Monitoring in Argentina (PuMA) collaboration systematically monitors southern glitching pulsars, maintaining high-cadence single-pulse records of the Vela pulsar. We present a pulse-per-pulse reanalysis of the 2024 major glitch of Vela (PSR~J0835$-$4510) with the 400~MHz-bandwidth ROACH backend of the Argentine Institute of Radioastronomy, and extend our machine-learning single-pulse pipeline---Isolation Forest outlier rejection, $\beta$-Variational-AutoEncoder denoising, and Self-Organizing-Map clustering---to new observations of PSR~J1644$-$4559 and the millisecond pulsar PSR~J0437$-$4715. For Vela, the 4- and 6-cluster decompositions of the eight days bracketing the glitch reproduce, with seven times the previous bandwidth, the behavior found with the narrow-band ETTUS receivers: higher-amplitude clusters peak earlier in phase, are narrower, more skewed, and less populated. With the glitch jump and its two exponential recovery terms included in the timing solution, the mean profile is stable to 1\% across all eight days (width change $-0.5\pm0.9$\% from pre- to post-glitch); omitting the recovery terms would mimic a post-glitch broadening of up to 55\% through a folding-frequency error at the $10^{-7}$ level. The clusters of PSR~J1644$-$4559 differ almost exclusively in amplitude, as expected for a scattering-dominated profile. For PSR~J0437$-$4715, retaining the 10\% of pulses with the highest peak-dominance score doubles the signal-to-noise ratio, and a five-cluster decomposition yields narrow, phase-ordered groups a factor $3.6\pm0.4$ narrower than the average profile, suggesting a $\sim$3.5-fold improvement in cluster-based timing precision for this pulsar-timing-array target, to be confirmed in a follow-up paper.

astro-ph.HE↗

arXiv2026

MeiBRD: Meta-Learning Intraoperative Biomechanical Residual Deformation

Accurate intraoperative liver registration is challenging due to substantial soft-tissue deformation yet sparse intraoperative measurements. Biomechanical models regularize this ill-posedness with prior knowledge but exhibit persistent prediction bias due to simplifying assumptions, while data-driven learning solutions struggle with data efficiency, generalization, and physical plausibility. We propose a hybrid registration framework that adapts a biomechanical prior using sparse intraoperative correspondences. Rather than learning a full deformation field, we learn a residual deformation function that corrects linear biomechanical predictions, modeled as a graph neural diffusion function with geometry-aware attention over the 3D liver mesh. To enable long-range information transfer of sparse observations, we take a novel perspective of sparse intraoperative measurements as \textit{context} samples where input-output pairs of the residual deformation function are fully observed, casting the problem into learning-to-learn this residual function from intraoperative context samples with feedforward meta-learners. Experiments on a deformable liver phantom dataset demonstrate improved registration accuracy and generalization compared to rigid, biomechanical, and data-driven baselines, particularly for out-of-distribution geometries and deformations.

cs.CV↗

arXiv2026

Learning Cardiac Electrophysiology Digital Twins Through Agentic Discovery of Hybrid Structure

Building personalized cardiac electrophysiology (EP) digital twins requires identifying the appropriate model structure for each patient, not merely fitting parameters. Traditional methods rely on experts to manually prescribe hybrid physics-neural architectures, which requires deep domain expertise and does not transfer across patients. Recent works have applied large language models (LLMs) to generate or act as hybrid models. However, despite their promising generalization capacity, these LLM-based methods lack the structural priors needed for stable cardiac simulations. Hence, we propose LEADS, a framework that formulates cardiac EP domain knowledge as a structured action space and utilizes an LLM agent to discover hybrid models. The agent follows an iterative reasoning-and-action loop to select, combine, and refine hybrid models, whilst gradient descent handles parameter fitting. The proposed LEADS designs every candidate model towards physically grounded, interpretable, and numerically stable, while allowing open-ended architectural discovery. We validate LEADS on synthetic data with three ground-truth reaction models and on real cardiac EP data, demonstrating that it outperforms both human-designed hybrid models and other LLM-based hybrid modeling.

cs.AI↗

arXiv2026

HAPI-EP: Towards Hybrid, Adaptive, and Predictive Digital Twins of Cardiac Electrophysiology

A digital twin (DT) of a patient-specific heart offers significant potential in personalized medicine. However, its rapid and dynamic adaptation to an individual's live data and its predictive capability after adaptation remains central challenges. We examine this challenge from its two building blocks: DT formulation where mechanistic and data-driven models show competing merits and limitations, and DT optimization strategies that are largely driven by a reconstruction objective leading to un-identifiable models. We address both bottlenecks via HAPI -- an AI framework for building hybrid, adaptive, and predictive DTs with three key enablers. First, HAPI constructs a physics-integrated gray-box model in which an interpretable mechanistic backbone is augmented by a neural component that models its residual to the observed data. Second, rather than attempting to pre-encode all possible variations in a static hybrid model, HAPI enables rapid on-the-fly adaptation of the hybrid model to few-shot live data, achieved by feedforward meta-learners realizing amortized inference of both mechanistic and neural parameters of the hybrid model trained with predictive objectives. Finally, we show that this adaptivity corresponds to the construction of a conditional generative model (i.e., the hybrid DT) that endows it with theoretical identifiability and thus strong performance in predictive scenarios. We demonstrate the proof-of-concept of HAPI in cardiac electrophysiology using a hybrid monodomain model with mechanistic reaction kinetics and neural graph diffusion. Across synthetic and real-data studies, we show that HAPI's mechanistic-neural hybridization and predictive adaptation are critical for obtaining identifiable DTs with strong predictive and out-of-distribution capabilities.

cs.LG↗

arXiv2026

cAPM: Continual AI-Assisted Pace-Mapping with Active Learning

Ventricular tachycardia is a life-threatening rhythm disorder and a major cause of sudden cardiac death. Pace-mapping is a clinical procedure for identifying the intervention target during catheter ablation of VT. It requires clinicians to pace different sites in the ventricles and rapidly interpret the resulting electrocardiograms to determine where to pace next or whether a target site has been identified. Active learning AI models have been proposed to guide clinicians to the next pacing site, showing promise in reducing the number of pacing sites and improving the efficiency of pace-mapping. Existing methods require retraining each target without the ability to transfer knowledge across multiple VTs within the same patient or across patients. We introduce cAPM for continuous AI-assisted pace-mapping to capture and transfer knowledge accumulated from past pace-mapping data to reduce the number of pace-mapping data needed for future target VTs. This is made possible by a task-agnostic surrogate neural network that learns the mapping from pacing sites to 12-lead ECG morphology, an active-learning strategy that refines this surrogate model by selecting the most informative pacing site for each target, and a continual learning strategy to do so sequentially while retaining knowledge from prior targets. Evaluated on an in-silico testbed consisting of sequentially-presented localization tasks across different physiological conditions and ventricular geometries, cAPM with and without replay of past data samples achieved an 81% probability of localizing within clinical tolerance (5 mm accuracy) using 4.5 pace-mapping sites, compared to the state-of-the-art active-learning method achieving 38% probability using 13.7 pacing sites. These results provide a strong basis for preparing cAPM towards in-vivo preclinical and clinical studies where it can be used to guide pace-mapping.

cs.LG↗

arXiv2026

CoMetaPNS: Continually Meta-learning Personalized Neural Surrogates for Cardiac Electrophysiology Simulations

Personalized virtual heart simulations face challenges in model personalization and computational cost. While neural surrogates offer state-of-the-art solutions, they typically address either efficient personalization or training generalizable models. Recent work reframes this by learning the process of personalizing a surrogate using limited subject-specific context data, through few-shot generative modeling with set-conditioned surrogates and meta-learned amortized inference. These methods, however, assume a static and diverse training distribution with known task identifiers. When new data becomes available, they require costly retraining with all prior data to avoid catastrophic forgetting - a phenomena where the model forgets earlier tasks when trained on new ones. This is a major limitation in clinical settings where often unlabeled data arrives sequentially and full retraining is infeasible. This paper presents a new continual meta-learning framework to achieve personalized neural surrogates able to not only continually integrate information but also identify whether incoming data stems from a known or unknown dynamics source. By leveraging a continual Bayesian Gaussian Mixture Model over a memory buffer, our framework can infer the identifiers and relationships of data over time - required for effective meta-learning. Empirical results on synthetic cardiac data demonstrate superior simulation forecasting, computational scalability, and resilience to catastrophic forgetting compared to existing baselines.

cs.LG↗

arXiv2025

Study of the 2024 major Vela glitch at the Argentine Institute of Radioastronomy

We report here on new results of the systematic monitoring of southern glitching pulsars at the Argentine Institute of Radioastronomy. In particular, we study in this work the new major glitch in the Vela pulsar (PSR J0835$-$4510) that occurred on 2024 April 29. We aim to thoroughly characterise the rotational behaviour of the Vela pulsar around its last major glitch and investigate the statistical properties of its individual pulses around the glitch. We characterise the rotational behaviour of the pulsar around the glitch through the pulsar timing technique. We measured the glitch parameters by fitting timing residuals to the data collected during the days surrounding the event. In addition, we study Vela individual pulses during the days of observation just before and after the glitch. We selected nine days of observations around the major glitch on 2024 April 29 and studied their statistical properties with the Self-Organizing Maps (SOM) technique. We used Variational AutoEncoder (VAE) reconstruction of the pulses to separate them clearly from the noise. We obtain a precise timing solution for the glitch. We find two recovery terms of $\sim 3~\mathrm{days}$ and $\sim 17~\mathrm{days}$. We find a correlation of high amplitude with narrower pulses while not finding notable qualitative systematic changes before and after the glitch.

astro-ph.HE↗

arXiv2024

Feasibility Study on Active Learning of Smart Surrogates for Scientific Simulations

High-performance scientific simulations, important for comprehension of complex systems, encounter computational challenges especially when exploring extensive parameter spaces. There has been an increasing interest in developing deep neural networks (DNNs) as surrogate models capable of accelerating the simulations. However, existing approaches for training these DNN surrogates rely on extensive simulation data which are heuristically selected and generated with expensive computation -- a challenge under-explored in the literature. In this paper, we investigate the potential of incorporating active learning into DNN surrogate training. This allows intelligent and objective selection of training simulations, reducing the need to generate extensive simulation data as well as the dependency of the performance of DNN surrogates on pre-defined training simulations. In the problem context of constructing DNN surrogates for diffusion equations with sources, we examine the efficacy of diversity- and uncertainty-based strategies for selecting training simulations, considering two different DNN architecture. The results set the groundwork for developing the high-performance computing infrastructure for Smart Surrogates that supports on-the-fly generation of simulation data steered by active learning strategies to potentially improve the efficiency of scientific simulations.

cs.LG↗

arXiv2024

STLLaVA-Med: Self-Training Large Language and Vision Assistant for Medical Question-Answering

Large Vision-Language Models (LVLMs) have shown significant potential in assisting medical diagnosis by leveraging extensive biomedical datasets. However, the advancement of medical image understanding and reasoning critically depends on building high-quality visual instruction data, which is costly and labor-intensive to obtain, particularly in the medical domain. To mitigate this data-starving issue, we introduce Self-Training Large Language and Vision Assistant for Medicine (STLLaVA-Med). The proposed method is designed to train a policy model (an LVLM) capable of auto-generating medical visual instruction data to improve data efficiency, guided through Direct Preference Optimization (DPO). Specifically, a more powerful and larger LVLM (e.g., GPT-4o) is involved as a biomedical expert to oversee the DPO fine-tuning process on the auto-generated data, encouraging the policy model to align efficiently with human preferences. We validate the efficacy and data efficiency of STLLaVA-Med across three major medical Visual Question Answering (VQA) benchmarks, demonstrating competitive zero-shot performance with the utilization of only 9% of the medical data.

cs.CV↗

arXiv2024

HyPer-EP: Meta-Learning Hybrid Personalized Models for Cardiac Electrophysiology

Personalized virtual heart models have demonstrated increasing potential for clinical use, although the estimation of their parameters given patient-specific data remain a challenge. Traditional physics-based modeling approaches are computationally costly and often neglect the inherent structural errors in these models due to model simplifications and assumptions. Modern deep learning approaches, on the other hand, rely heavily on data supervision and lacks interpretability. In this paper, we present a novel hybrid modeling framework to describe a personalized cardiac digital twin as a combination of a physics-based known expression augmented by neural network modeling of its unknown gap to reality. We then present a novel meta-learning framework to enable the separate identification of both the physics-based and neural components in the hybrid model. We demonstrate the feasibility and generality of this hybrid modeling framework with two examples of instantiations and their proof-of-concept in synthetic experiments.

eess.SP↗

arXiv2024

Unsupervised Learning of Hybrid Latent Dynamics: A Learn-to-Identify Framework

Modern applications increasingly require unsupervised learning of latent dynamics from high-dimensional time-series. This presents a significant challenge of identifiability: many abstract latent representations may reconstruct observations, yet do they guarantee an adequate identification of the governing dynamics? This paper investigates this challenge from two angles: the use of physics inductive bias specific to the data being modeled, and a learn-to-identify strategy that separates forecasting objectives from the data used for the identification. We combine these two strategies in a novel framework for unsupervised meta-learning of hybrid latent dynamics (Meta-HyLaD) with: 1) a latent dynamic function that hybridize known mathematical expressions of prior physics with neural functions describing its unknown errors, and 2) a meta-learning formulation to learn to separately identify both components of the hybrid dynamics. Through extensive experiments on five physics and one biomedical systems, we provide strong evidence for the benefits of Meta-HyLaD to integrate rich prior knowledge while identifying their gap to observed data.

cs.LG↗

arXiv2024

Hybrid Kinetics Embedding Framework for Dynamic PET Reconstruction

In dynamic positron emission tomography (PET) reconstruction, the importance of leveraging the temporal dependence of the data has been well appreciated. Current deep-learning solutions can be categorized in two groups in the way the temporal dynamics is modeled: data-driven approaches use spatiotemporal neural networks to learn the temporal dynamics of tracer kinetics from data, which relies heavily on data supervision; physics-based approaches leverage \textit{a priori} tracer kinetic models to focus on inferring their parameters, which relies heavily on the accuracy of the prior kinetic model. In this paper, we marry the strengths of these two approaches in a hybrid kinetics embedding (HyKE-Net) framework for dynamic PET reconstruction. We first introduce a novel \textit{hybrid} model of tracer kinetics consisting of a physics-based function augmented by a neural component to account for its gap to data-generating tracer kinetics, both identifiable from data. We then embed this hybrid model at the latent space of an encoding-decoding framework to enable both supervised and unsupervised identification of the hybrid kinetics and thereby dynamic PET reconstruction. Through both phantom and real-data experiments, we demonstrate the benefits of HyKE-Net -- especially in unsupervised reconstructions -- over existing physics-based and data-driven baselines as well as its ablated formulations where the embedded tracer kinetics are purely physics-based, purely neural, or hybrid but with a non-adaptable neural component.

eess.IV↗

arXiv2024

LIBR+: Improving Intraoperative Liver Registration by Learning the Residual of Biomechanics-Based Deformable Registration

The surgical environment imposes unique challenges to the intraoperative registration of organ shapes to their preoperatively-imaged geometry. Biomechanical model-based registration remains popular, while deep learning solutions remain limited due to the sparsity and variability of intraoperative measurements and the limited ground-truth deformation of an organ that can be obtained during the surgery. In this paper, we propose a novel \textit{hybrid} registration approach that leverage a linearized iterative boundary reconstruction (LIBR) method based on linear elastic biomechanics, and use deep neural networks to learn its residual to the ground-truth deformation (LIBR+). We further formulate a dual-branch spline-residual graph convolutional neural network (SR-GCN) to assimilate information from sparse and variable intraoperative measurements and effectively propagate it through the geometry of the 3D organ. Experiments on a large intraoperative liver registration dataset demonstrated the consistent improvements achieved by LIBR+ in comparison to existing rigid, biomechnical model-based non-rigid, and deep-learning based non-rigid approaches to intraoperative liver registration.

eess.IV↗

arXiv2023

Distributionally Robust Optimization and Invariant Representation Learning for Addressing Subgroup Underrepresentation: Mechanisms and Limitations

Spurious correlation caused by subgroup underrepresentation has received increasing attention as a source of bias that can be perpetuated by deep neural networks (DNNs). Distributionally robust optimization has shown success in addressing this bias, although the underlying working mechanism mostly relies on upweighting under-performing samples as surrogates for those underrepresented in data. At the same time, while invariant representation learning has been a powerful choice for removing nuisance-sensitive features, it has been little considered in settings where spurious correlations are caused by significant underrepresentation of subgroups. In this paper, we take the first step to better understand and improve the mechanisms for debiasing spurious correlation due to subgroup underrepresentation in medical image classification. Through a comprehensive evaluation study, we first show that 1) generalized reweighting of under-performing samples can be problematic when bias is not the only cause for poor performance, while 2) naive invariant representation learning suffers from spurious correlations itself. We then present a novel approach that leverages robust optimization to facilitate the learning of invariant representations at the presence of spurious correlations. Finetuned classifiers utilizing such representation demonstrated improved abilities to reduce subgroup performance disparity, while maintaining high average and worst-group performance.

cs.CV↗

arXiv2023

Learning Transferable Object-Centric Diffeomorphic Transformations for Data Augmentation in Medical Image Segmentation

Obtaining labelled data in medical image segmentation is challenging due to the need for pixel-level annotations by experts. Recent works have shown that augmenting the object of interest with deformable transformations can help mitigate this challenge. However, these transformations have been learned globally for the image, limiting their transferability across datasets or applicability in problems where image alignment is difficult. While object-centric augmentations provide a great opportunity to overcome these issues, existing works are only focused on position and random transformations without considering shape variations of the objects. To this end, we propose a novel object-centric data augmentation model that is able to learn the shape variations for the objects of interest and augment the object in place without modifying the rest of the image. We demonstrated its effectiveness in improving kidney tumour segmentation when leveraging shape variations learned both from within the same dataset and transferred from external datasets.

cs.CV↗

arXiv2022

Interpretable Modeling and Reduction of Unknown Errors in Mechanistic Operators

Prior knowledge about the imaging physics provides a mechanistic forward operator that plays an important role in image reconstruction, although myriad sources of possible errors in the operator could negatively impact the reconstruction solutions. In this work, we propose to embed the traditional mechanistic forward operator inside a neural function, and focus on modeling and correcting its unknown errors in an interpretable manner. This is achieved by a conditional generative model that transforms a given mechanistic operator with unknown errors, arising from a latent space of self-organizing clusters of potential sources of error generation. Once learned, the generative model can be used in place of a fixed forward operator in any traditional optimization-based reconstruction process where, together with the inverse solution, the error in prior mechanistic forward operator can be minimized and the potential source of error uncovered. We apply the presented method to the reconstruction of heart electrical potential from body surface potential. In controlled simulation experiments and in-vivo real data experiments, we demonstrate that the presented method allowed reduction of errors in the physics-based forward operator and thereby delivered inverse reconstruction of heart-surface potential with increased accuracy.

eess.IV↗

arXiv2022

First results of the glitching pulsars monitoring program at the Argentine Institute of Radioastronomy

We report here on the first results of a systematic monitoring of southern glitching pulsars at the Argentine Institute of Radioastronomy that started in the year 2019. We detected a major glitch in the Vela pulsar (PSR J0835$-$4510) and two small-glitches in PSR J1048$-$5832. For each glitch, we present the measurement of glitch parameters by fitting timing residuals. We then make an individual pulses study of Vela in observations before and after the glitch. We selected 6 days of observations around the major glitch on 2021 July 22 and study their statistical properties with machine learning techniques. We use Variational AutoEncoder (VAE) reconstruction of the pulses to separate them clearly from the noise. We perform a study with Self-Organizing Maps (SOM) clustering techniques to search for unusual behavior of the clusters during the days around the glitch not finding notable qualitative changes. We have also detected and confirm recent glitches in PSR J0742$-$2822 and PSR J1740$-$3015.

astro-ph.HE↗

arXiv2022

Few-shot Generation of Personalized Neural Surrogates for Cardiac Simulation via Bayesian Meta-Learning

Clinical adoption of personalized virtual heart simulations faces challenges in model personalization and expensive computation. While an ideal solution is an efficient neural surrogate that at the same time is personalized to an individual subject, the state-of-the-art is either concerned with personalizing an expensive simulation model, or learning an efficient yet generic surrogate. This paper presents a completely new concept to achieve personalized neural surrogates in a single coherent framework of meta-learning (metaPNS). Instead of learning a single neural surrogate, we pursue the process of learning a personalized neural surrogate using a small amount of context data from a subject, in a novel formulation of few-shot generative modeling underpinned by: 1) a set-conditioned neural surrogate for cardiac simulation that, conditioned on subject-specific context data, learns to generate query simulations not included in the context set, and 2) a meta-model of amortized variational inference that learns to condition the neural surrogate via simple feed-forward embedding of context data. As test time, metaPNS delivers a personalized neural surrogate by fast feed-forward embedding of a small and flexible number of data available from an individual, achieving -- for the first time -- personalization and surrogate construction for expensive simulations in one end-to-end learning framework. Synthetic and real-data experiments demonstrated that metaPNS was able to improve personalization and predictive accuracy in comparison to conventionally-optimized cardiac simulation models, at a fraction of computation.

cs.LG↗