VerIde ECG Biometrics: Verification and Identification

This work studies electrocardiogram (ECG) biometrics at large scale, evaluating how strongly an ECG can be linked to an individual and, consequently, how its anonymization may be compromised. We show that identity information is already present in tabular representations (fiducial features): even a simple MLP-based embedding network yields non-trivial performance, indicating that anonymization based solely on releasing features does not guarantee privacy. We then adopt embedding-based deep learning models (ArcFace), first on features and then on ECG waveforms, showing a performance jump when moving from tabular inputs to waveforms, and a further gain with larger training sets and consistent normalization across train/val/test. On a large-scale test set, verification achieves high TAR at strict FAR thresholds (TAR=0.908 @ FAR=1e-3; TAR=0.820 @ FAR=1e-4) with EER=2.53% (all-vs-all); closed-set identification yields Rank@1=0.812 and Rank@10=0.910. In open-set, a two-stage pipeline (top-K shortlist on embeddings + re-ranking) reaches DIR@FAR up to 0.976 at FAR=1e-3 and 1e-4. Overall, the results show that ECG carries a measurable individual signature: re-identification is already possible with tabular features and is further amplified by embedding-based models, making privacy implications and realistic operational protocols essential to consider.

Metric Hub: A metric library and practical selection workflow for use-case-driven data quality assessment in medical AI

Machine learning (ML) in medicine has transitioned from research to concrete applications aimed at supporting several medical purposes like therapy selection, monitoring and treatment. Acceptance and effective adoption by clinicians and patients, as well as regulatory approval, require evidence of trustworthiness. A major factor for the development of trustworthy AI is the quantification of data quality for AI model training and testing. We have recently proposed the METRIC-framework for systematically evaluating the suitability (fit-for-purpose) of data for medical ML for a given task. Here, we operationalize this theoretical framework by introducing a collection of data quality metrics - the metric library - for practically measuring data quality dimensions. For each metric, we provide a metric card with the most important information, including definition, applicability, examples, pitfalls and recommendations, to support the understanding and implementation of these metrics. Furthermore, we discuss strategies and provide decision trees for choosing an appropriate set of data quality metrics from the metric library given specific use cases. We demonstrate the impact of our approach exemplarily on the PTB-XL ECG-dataset. This is a first step to enable fit-for-purpose evaluation of training and test data in practice as the base for establishing trustworthy AI in medicine.

Looking Beyond Accuracy: A Holistic Benchmark of ECG Foundation Models

The electrocardiogram (ECG) is a cost-effective, highly accessible and widely employed diagnostic tool. With the advent of Foundation Models (FMs), the field of AI-assisted ECG interpretation has begun to evolve, as they enable model reuse across different tasks by relying on embeddings. However, to responsibly employ FMs, it is crucial to rigorously assess to which extent the embeddings they produce are generalizable, particularly in error-sensitive domains such as healthcare. Although prior works have already addressed the problem of benchmarking ECG-expert FMs, they focus predominantly on the evaluation of downstream performance. To fill this gap, this study aims to find an in-depth, comprehensive benchmarking framework for FMs, with a specific focus on ECG-expert ones. To this aim, we introduce a benchmark methodology that complements performance-based evaluation with representation-level analysis, leveraging SHAP and UMAP techniques. Furthermore, we rely on the methodology for carrying out an extensive evaluation of several ECG-expert FMs pretrained via state-of-the-art techniques over different cross-continental datasets and data availability settings; this includes ones featuring data scarcity, a fairly common situation in real-world medical scenarios. Experimental results show that our benchmarking protocol provides a rich insight of ECG-expert FMs' embedded patterns, enabling a deeper understanding of their representational structure and generalizability.

ECG-Agent: On-Device Tool-Calling Agent for ECG Multi-Turn Dialogue

Recent advances in Multimodal Large Language Models have rapidly expanded to electrocardiograms, focusing on classification, report generation, and single-turn QA tasks. However, these models fall short in real-world scenarios, lacking multi-turn conversational ability, on-device efficiency, and precise understanding of ECG measurements such as the PQRST intervals. To address these limitations, we introduce ECG-Agent, the first LLM-based tool-calling agent for multi-turn ECG dialogue. To facilitate its development and evaluation, we also present ECG-Multi-Turn-Dialogue (ECG-MTD) dataset, a collection of realistic user-assistant multi-turn dialogues for diverse ECG lead configurations. We develop ECG-Agents in various sizes, from on-device capable to larger agents. Experimental results show that ECG-Agents outperform baseline ECG-LLMs in response accuracy. Furthermore, on-device agents achieve comparable performance to larger agents in various evaluations that assess response accuracy, tool-calling ability, and hallucinations, demonstrating their viability for real-world applications.

ECGFlowCMR: Pretraining with ECG-Generated Cine CMR Improves Cardiac Disease Classification and Phenotype Prediction

Cardiac Magnetic Resonance (CMR) imaging provides a comprehensive assessment of cardiac structure and function but remains constrained by high acquisition costs and reliance on expert annotations, limiting the availability of large-scale labeled datasets. In contrast, electrocardiograms (ECGs) are inexpensive, widely accessible, and offer a promising modality for conditioning the generative synthesis of cine CMR. To this end, we propose ECGFlowCMR, a novel ECG-to-CMR generative framework that integrates a Phase-Aware Masked Autoencoder (PA-MAE) and an Anatomy-Motion Disentangled Flow (AMDF) to address two fundamental challenges: (1) the cross-modal temporal mismatch between multi-beat ECG recordings and single-cycle CMR sequences, and (2) the anatomical observability gap due to the limited structural information inherent in ECGs. Extensive experiments on the UK Biobank and a proprietary clinical dataset demonstrate that ECGFlowCMR can generate realistic cine CMR sequences from ECG inputs, enabling scalable pretraining and improving performance on downstream cardiac disease classification and phenotype prediction tasks.

Physics-Integrated Inference for Signal Recovery in Non-Gaussian Regimes

High-performance room-temperature sensing is often limited by non-stationary $1/f$ fluctuations and non-Gaussian stochasticity. In spintronic devices, thermally activated Néel switching creates heavy-tailed noise that masks weak signals, defeating linear filters optimized for Gaussian statistics. Here, we introduce a physics-integrated inference framework that decouples signal morphology from stochastic transients using a hierarchical 1D CNN-GRU topology. By learning the temporal signatures of Néel relaxation, this architecture reduces the Noise Equivalent Differential Temperature (NEDT) of spintronic Poisson bolometers by a factor of six (233.78 mK to 40.44 mK), effectively elevating room-temperature sensitivity toward cryogenic limits. We demonstrate the framework's universality across the electromagnetic and biological spectrum, achieving a 9-fold error suppression in Radar tracking, a 40\% uncertainty reduction in LiDAR, and a 15.56 dB SNR enhancement in ECG. This hardware-inference coupling recovers deterministic signals from fluctuation-dominated regimes, enabling near-ideal detection limits in noisy edge environments.

How Much Temporal Modeling is Enough? A Systematic Study of Hybrid CNN-RNN Architectures for Multi-Label ECG Classification

Accurate multi-label classification of electrocardiogram (ECG) signals remains challenging due to the coexistence of multiple cardiac conditions, pronounced class imbalance, and long-range temporal dependencies in multi-lead recordings. Although recent studies increasingly rely on deep and stacked recurrent architectures, the necessity and clinical justification of such architectural complexity have not been rigorously examined. In this work, we perform a systematic comparative evaluation of convolutional neural networks (CNNs) combined with multiple recurrent configurations, including LSTM, GRU, Bidirectional LSTM (BiLSTM), and their stacked variants, for multi-label ECG classification on the PTB-XL dataset comprising 23 diagnostic categories. The CNN component serves as a morphology-driven baseline, while recurrent layers are progressively integrated to assess their contribution to temporal modeling and generalization performance. Experimental results indicate that a CNN integrated with a single BiLSTM layer achieves the most favorable trade-off between predictive performance and model complexity. This configuration attains superior Hamming loss (0.0338), macro-AUPRC (0.4715), micro-F1 score (0.6979), and subset accuracy (0.5723) compared with deeper recurrent combinations. Although stacked recurrent models occasionally improve recall for specific rare classes, our results provide empirical evidence that increasing recurrent depth yields diminishing returns and may degrade generalization due to reduced precision and overfitting. These findings suggest that architectural alignment with the intrinsic temporal structure of ECG signals, rather than increased recurrent depth, is a key determinant of robust performance and clinically relevant deployment.

UniPACT: A Multimodal Framework for Prognostic Question Answering on Raw ECG and Structured EHR

Accurate clinical prognosis requires synthesizing structured Electronic Health Records (EHRs) with real-time physiological signals like the Electrocardiogram (ECG). Large Language Models (LLMs) offer a powerful reasoning engine for this task but struggle to natively process these heterogeneous, non-textual data types. To address this, we propose UniPACT (Unified Prognostic Question Answering for Clinical Time-series), a unified framework for prognostic question answering that bridges this modality gap. UniPACT's core contribution is a structured prompting mechanism that converts numerical EHR data into semantically rich text. This textualized patient context is then fused with representations learned directly from raw ECG waveforms, enabling an LLM to reason over both modalities holistically. We evaluate UniPACT on the comprehensive MDS-ED benchmark, it achieves a state-of-the-art mean AUROC of 89.37% across a diverse set of prognostic tasks including diagnosis, deterioration, ICU admission, and mortality, outperforming specialized baselines. Further analysis demonstrates that our multimodal, multi-task approach is critical for performance and provides robustness in missing data scenarios.

Adaptive Non-Uniform Sampling of Bandlimited Signals via Algorithm-Encoder Co-Design

We propose an adaptive non-uniform sampling framework for bandlimited signals based on an algorithm-encoder co-design perspective. By revisiting the convergence analysis of iterative reconstruction algorithms for non-uniform measurements, we derive a local, energy-based sufficient condition that governs reconstruction behavior as a function of the signal and derivative energies within each sampling interval. Unlike classical approaches that impose a global Nyquist-type bound on the inter-sample spacing, the proposed condition permits large gaps in slowly varying regions while enforcing denser sampling only where the signal exhibits rapid temporal variation. Building on this theoretical insight, we design a variable-bias, variable-threshold integrate-and-fire time encoding machine (VBT-IF-TEM) whose firing mechanism is explicitly shaped to enforce the derived local convergence condition. To ensure robustness, a shifted-signal formulation is introduced to suppress excessive firing in regions where the magnitude of the signal amplitude is close to zero or the local signal energy approaches zero. Using the proposed encoder, an analog signal is discretely represented by time encodings and signal averages, enabling perfect reconstruction via a standard iterative algorithm even when the local sampling rate falls below the Nyquist rate. Simulation results on synthetic signals and experiments on ultrasonic guided-wave and ECG signals demonstrate that the proposed framework achieves substantial reductions in sampling density compared to uniform sampling and conventional IF-TEMs, while maintaining accurate reconstruction. The results further highlight a controllable tradeoff between sampling density, reconstruction accuracy, and convergence behavior, which can be navigated through adaptive parameter selection.

Beat-ssl: Capturing Local ECG Morphology through Heartbeat-level Contrastive Learning with Soft Targets

Obtaining labelled ECG data for developing supervised models is challenging. Contrastive learning (CL) has emerged as a promising pretraining approach that enables effective transfer learning with limited labelled data. However, existing CL frameworks either focus solely on global context or fail to exploit ECG-specific characteristics. Furthermore, these methods rely on hard contrastive targets, which may not adequately capture the continuous nature of feature similarity in ECG signals. In this paper, we propose Beat-SSL, a contrastive learning framework that performs dual-context learning through both rhythm-level and heartbeat-level contrasting with soft targets. We evaluated our pretrained model on two downstream tasks: 1) multilabel classification for global rhythm assessment, and 2) ECG segmentation to assess its capacity to learn representations across both contexts. We conducted an ablation study and compared the best configuration with three other methods, including one ECG foundation model. Despite the foundation model's broader pretraining, Beat-SSL reached 93% of its performance in multilabel classification task and surpassed all other methods in the segmentation task by 4%.