Physiological Measurement

We examine the long-range temporal structure of forecasts produced by Time-Series Foundation Models (TSFMs) on heartbeat dynamics using the MIT-BIH Normal Sinus Rhythm Database (NSRDB). Our findings indicate that these models do not adequately capture long-range dependencies, as reflected in growing errors in RR-interval predictions over longer forecast horizons. Code is available at https://github.com/SubramanianLab/ecg-tsfm-benchmark.

High-frequency physiological monitoring in ICUs can identify impending deterioration hours before clinical recognition yet extracting reliable early-warning signals from noisy vital-sign streams remains challenging. We present SIgnose, an interpretable prediction framework for early detection of abnormal shock index (SI), built from routinely monitored vital signs using physiologic variability and nonlinear time-series features. SIgnose was developed on the eICU Collaborative Research Database and externally validated on the MIMIC-III adult database and a pediatric SafeICU cohort (AIIMS New Delhi), with additional prospective validation in the pediatric ICU. We benchmarked three representation strategies: (i) engineered physiologic variability and nonlinear time-series features, (ii) deep learning, and (iii) Llama-3.1-8B embeddings with low-rank adaptation. Physiologic variability features consistently demonstrated superior cross-cohort generalization. The final model used 3,970 features from five vital signs to predict abnormal SI up to 8 hours ahead, achieving AUROC 0.861 (95% CI 0.859-0.863) and AUPRC 0.927 (95% CI 0.925-0.929) on eICU. External validation yielded AUROC 0.870 (95% CI 0.863-0.876) and AUPRC 0.935 (95% CI 0.930-0.940) on MIMIC-III, and AUROC 0.875 (95% CI 0.863-0.888) and AUPRC 0.915 (95% CI 0.898-0.930) on SafeICU; prospective pediatric validation (n = 88) achieved AUROC 0.885 (95% CI 0.868-0.902) and AUPRC 0.911 (95% CI 0.882-0.936). SHAP interpretability analysis identified heart rate variability, respiratory trend dynamics, and multi-scale blood pressure variability as key early-warning signatures. These findings establish SIgnose as a reproducible, low-compute, early-warning framework and demonstrate that physiologic variability features provide robust, generalizable representations for early deterioration detection across adult and pediatric critical care.

Non-invasive wrist pulse monitoring has been integrated into various medical systems for cardiovascular assessment. However, different definitions of pulse transit time are used in the literature, and their statistical behavior when measured locally at the wrist using pressure sensors has not been systematically examined. Wearable wristbands designed to measure pulse transit time (PTT) have emerged as valuable tools for evaluating cardiac activity. While several algorithms have been developed to predict blood pressure using PTT, it is well recognized that PTT and its inverse parameter, pulse wave velocity (PWV), exhibit temporal variability. In this study, PTT was explicitly measured at the wrist's radial artery to investigate its statistical variation and relationship with different arterial pressures. The experiment exhibits two distinct methodologies for PTT computation using onset-based and peak based measurements. Data were recorded across five cuff pressure levels at 20, 40, 60, 80, and 100 mmHg using the pulse pressure sensor (PPS). PTTonset time shows lower coefficient of variation as compared to PTTpeak time within the 100 mmHg pressure range. The weak correlation coefficient is recorded between PTT values. However, dynamic time warping (DTW) analysis revealed a notable similarity in the time series of PTTonset and PTTpeak, regardless of the applied pressure level. For the multi participant dataset, the mean DTW distances ranged from 0.029 to 0.046 across the tested cuff pressures, illustrating consistent similarity between PTTonset and PTTpeak over time. The objective of this study is to examine the statistical behavior, stability, and temporal similarity of the two commonly used PTT definitions when measured at the radial artery using pressure sensors. Statistical analysis shows consistent differences between the two PTT definitions across participants. PTTonset shows lower variation than PTTpeak. However, PTTpeak requires simpler computation and produces fewer detection errors, while PTTonset provides lower statistical variation.

Ventricular tachycardia (VT) and ventricular fibrillation (VF) are the leading electrical causes of sudden cardiac death, but automated detection is limited by strong class imbalance, where lethal arrhythmias account for fewer than 22% of ECG segments. In this setting, standard classifiers can achieve high accuracy by predicting normal rhythm in most cases while missing many lethal events, a failure mode referred to as rare-class collapse. We evaluated six imbalance-handling approaches: naive logistic regression, inverse-frequency reweighting, label-distribution-aware margin loss (LDAM), cost-sensitive training, two-stage cascade classification, and anomaly detection on 15,614 ECG segments from three PhysioNet databases (VTaC, VFDB, CUDB), with an overall normal-to-lethal ratio of 3.6:1. All methods were assessed at a fixed operating point of 95% specificity using recall, area under the precision-recall curve (AUPRC), and missed-lethal-event rate (MLER). The naive model achieved 45.1% recall (MLER = 0.549), missing 564 of 1,027 lethal events despite 84.1% accuracy. The two-stage cascade performed best, with 65.2% recall, AUPRC of 0.821, and MLER of 0.348, reducing missed events by 37% and achieving the highest decision-curve net benefit. Per-source analysis showed near-complete VF detection (recall up to 0.975) but much lower VT detection (recall 0.183), suggesting a feature-space limitation due to spectral similarity between organized VT and rapid sinus rhythm. Overall, the results show that evaluation metrics strongly influence the visibility of rare-class failure, and that cascade-based methods outperform simpler reweighting approaches for detecting lethal arrhythmias.

Background Continuous health monitoring enables early detection of diseases and improves therapeutic outcomes. Non-intrusive biosignal sensors, such as capacitive ECG (cECG), offer a practical solution for daily monitoring in private environments, such as smart homes and vehicles. However, artifacts reduce signal quality and compromise reliability. Methods Following a registered report protocol (Warnecke JM et al. Plos One. 2021; 16(7):e0254780), we record data of 44 subjects and develop an artifact index for cECG. We use three signal quality indices (SQIs): the correlation of QRS complexes (corSQI), the R-peak detection consistency (bSQI) and the absolute amplitude ratio (aSQI). Our index classifies overlapping 10s segments with a step-width of 2s into clean or artifact segments. We label a 2s interval as artifacts if all five overlapping segments indicate artifacts. We record cECGs using an armchair with integrated electrodes in a single-arm study involving 44 subjects performing two activities -- reading and watching television (TV); for 11 minutes each. We record a time-synchronized reference ECG with skin electrodes on the chest. To evaluate the artifact index, we compare it with manually generated ground truth. Moreover, we evaluate the clothing materials cotton, linen, jeans, and polyester in 5 subjects. Results Watching TV results in longer, continuously clean signal durations than reading. On average, 88.3% of the signal has a minimum continuous clean duration of 10s, versus 79.8% during reading. All clothing configurations achieve a clean signal duration exceeding 10s. Among the SQI metrics, bSQI performs best, achieving an accuracy of 90.7% and an F1 score of 79.9%. Combining the three SQI metrics in a voting approach improves accuracy to 92.0% and F1 score to 82.1%. Discussion Our artifact index automatically distinguishes clean from artifact cECG segments, promoting health monitoring in unsupervised real-world settings, earlier disease detection, and preventive health management. A limitation is the investigation of only two scenarios (reading and watching TV).

Purpose: Continuous intracranial pressure (ICP) monitoring requires invasive instrumentation, reaching only a narrow subset of critically ill patients. We tested whether deep learning models trained on routinely acquired extracranial signals can reconstruct continuous ICP waveforms at clinically relevant accuracy in an independent external cohort. Methods: In adults admitted to the ICU at a single quaternary health system, five deep learning architectures were trained on high-frequency arterial blood pressure (ABP), photoplethysmography (PPG), and electrocardiography (ECG) waveforms, using invasive (intraparenchymal) ICP as ground truth. Two fusion strategies (early and late) and three training objectives (waveform-morphology, baseline robust regression, and weighted robust regression) were evaluated. Models were externally validated on the held-out MIMIC-III Waveform Database. Performance was assessed by mean absolute error (MAE) and waveform similarity by Pearson correlation (r). Results: We analyzed data from 158 critically ill adults (~5,322 hours) across two quaternary health systems (Johns Hopkins Hospital, Baltimore; Beth Israel Deaconess Medical Center, Boston). Validation MAE ranged from 4.276 mmHg [95% CI 4.269, 4.283] (gated recurrent, late fusion) to 4.946 mmHg [95% CI 4.938, 4.956] (attention-based, early fusion), with Pearson r ranging from 0.599 [95% CI 0.599, 0.600] to 0.722 [95% CI 0.722, 0.723]. The multiscale encoder-decoder model demonstrated the most favorable MAE-correlation tradeoff. Conclusion: This is the first demonstration that continuous ICP waveform reconstruction from bedside signals generalizes across institutions at clinically relevant accuracy, establishing a foundation for non-invasive ICP monitoring and motivating validation across broader populations and ICP ranges.

Continuous monitoring of Arterial Blood Pressure (ABP) in critically ill patients requires invasive arterial catheterization, which carries risks of thrombosis, vascular injury and infection. Here, we train and validate a computational model for continuous non-invasive ABP estimation in Intensive Care Unit (ICU) patients using a novel wearable sensor array. The sensor acquires continuous high frequency photoplethysmography (PPG) and electrocardiography (ECG) signals which are used as inputs in a deep learning algorithm for beat-to-beat reconstruction of ABP waveforms. We include 28 patients enrolled in four ICU units at Johns Hopkins Hospital, comprising 15,489 five-second ECG and PPG segments. A CNN/LSTM hybrid architecture achieved an R2 of 0.812 and a sample-level mean absolute error (MAE) of 4.94 {+/-} 4.96 mmHg, with systolic and diastolic blood pressure MAEs of 6.38 {+/-} 6.62 and 3.99 {+/-} 4.53 mmHg, respectively. This performance closely approached an upper-bound model trained on contemporaneously acquired ground truth ECG and PPG signals (R2 = 0.824, MAE = 4.81 mmHg), indicating that the sensors retain most hemodynamically relevant information. Split-conformal prediction provided calibrated uncertainty intervals with coverage meeting nominal targets, offering a principled framework for bedside confidence assessment. These findings demonstrate the feasibility of accurate, continuous, non-invasive ABP waveform estimation from wearable biosignals in critically ill patients, establishing a foundation for reducing dependence on invasive arterial monitoring while preserving the waveform-level information essential for hemodynamic management.

Deep-learning ECG analysis is advancing rapidly but lacks stable, physiologically interpretable indicators to anchor explainable artificial intelligence (AI). Tensor cardiography (TCG) models electrocardiographic (ECG) waveforms as differences between pairs of cumulative distribution functions (CDFs), representing collective myocardial action potential transitions. However, the original 4-CDF model has limitations in fitting P waves and complex QRST patterns. This study aimed to evaluate whether increasing the number of CDFs from 4 to 10 improves TCG fitting accuracy and to characterize normative distributions of 10-CDF parameters in healthy individuals. Participants were recruited through occupational health screening at Tobu Railway Co., Ltd. (n = 415) and from the Nippon Medical School Hospital ECG database (n = 29). Standard 12-lead ECGs from 444 healthy participants, including 345 men and 99 women with a mean age of 46.9 years, were analyzed using TCG software. Reconstruction accuracy was assessed using RMSE, paired t-tests, and Cohens d. The 10-CDF model achieved significantly lower RMSE values across all leads than the 4-CDF model, with all p values < 0.0001 and very large effect sizes. In representative leads, RMSEs for the 4-CDF versus 10-CDF models were 0.0256 versus 0.0061 in lead II, 0.0230 versus 0.0063 in lead V1, and 0.0265 versus 0.0062 in lead V5. The coefficient of determination improved from a median of 0.952 with the 4-CDF model to 0.997 with the 10-CDF model in lead II. Parameter dispersion was reduced, suggesting improved estimation stability. Two new parameters, T_mean_diff and RT_mean_duration, were derivable from the expanded model; RT_mean_duration showed significant correlations with age and body surface area. In conclusion, increasing the CDF resolution from 4 to 10 significantly enhanced ECG waveform reconstruction accuracy and parameter stability. These findings provide normative distributions of 10-CDF TCG parameters and may support future explainable AI-based ECG analysis.

Adhesive electrocardiography (ECG) electrodes used in neonatal intensive care units (NICUs) may cause skin injury in premature infants. Although photoplethysmography (PPG)-based ECG reconstruction has been explored, existing studies have mainly focused on adult data and often rely on direct PPG-to-ECG mapping or artificial signal alignment, which may be unsuitable for neonates with highly variable pulse arrival time (PAT). In this study, we propose an alignment-free RoPE-based dual-stream Transformer for reconstructing missing neonatal ECG segments using concurrent PPG signals and bidirectional ECG context. A total of 52,566 10-second ECG-PPG windows were extracted from 159 NICU patients and split at the patient level to prevent data leakage. The model was designed to learn ECG-PPG temporal coupling without forced synchronization by integrating PPG-derived hemodynamic timing information with lead-specific ECG context. Under a 40% random missing condition, the model achieved a Pearson correlation coefficient of 0.96, mean absolute error of 0.04, and root mean square error of 0.07. It also maintained robust performance under 4.0-second continuous block loss and 60% random patch loss, preserving a PCC of at least 0.90. These findings suggest that the proposed framework may serve as a signal imputation module for maintaining ECG monitoring continuity in NICU environments. Prospective validation is required before clinical diagnostic use.

Consumer wearable devices increasingly use single-lead electrocardiograms (ECGs) for cardiac monitoring, but these signals contain substantially less spatial information than the clinical 12-lead standard. Whether this reduction dispro- portionately affects older adults, who often present with more complex cardiac conditions, remains poorly understood. In this study, we evaluated the impact of lead reduction on AI-ECG diagnostic performance across age groups. A 1D resid- ual neural network was trained on 21,091 PTB-XL ECG recordings spanning five diagnostic superclasses and assessed using 12-, 6-, 2-, and 1-lead configurations. Under the full 12-lead setting, model accuracy declined from 84.5% in patients younger than 40 years to 66.2% in patients aged 75 years or older. Progressive lead reduction further widened this gap. Under the 1-lead configuration, accuracy decreased by 14.1 percentage points in the 75+ group but by only 0.4 percent- age points in the <40 group, representing an approximately 40-fold differential degradation confirmed by three independent statistical tests (all p < 0.0001). Older adults also exhibited greater multi-condition diagnostic complexity, pro- viding a plausible explanation for their increased vulnerability to information loss. External validation on the MIT-BIH Arrhythmia Database confirmed cross- dataset model stability. These findings suggest that age-stratified performance reporting should be a minimum standard in wearable AI-ECG validation and regulatory assessment.

Objective: Patient-ventilator dyssynchrony (PVD) is a common and clinically consequential problem in critically ill patients receiving invasive mechanical ventilation. Yet automated identification of PVD subtypes at scale remains an unmet clinical need, owing to the lack of large annotated bedside waveform datasets. Methods: We developed and validated a semi-supervised algorithm for automated annotation of PVD. In two medical ICUs at a tertiary academic center, bedside devices continuously collected airway flow and pressure waveforms from the ventilators. We developed a software interface with an information retrieval system that grouped similar breaths for expert human review, yielding 1,542,296 labeled breaths across eight categories: 2 labels for breath delivery mode, 5 labels for PVD subtypes, and 1 label denoting a normal breath. Two pulmonary physicians with expertise in ventilator training and education provided the expert reference labels. We trained an initial classification model on a model-derivation set of 771,148 breaths (divided into training and validation) and evaluated it on a hold-out test set of 771,149 breaths A semi-supervised approach was utilized to extend labeling to an additional 12,965,000 unlabeled breaths. Results: The supervised model performed well across all labels, with Macro-F1 scores between 0.96 and 1.00. Semi-supervised learning across 12 rounds expanded the training set from 771,148 to 8,563,995 breaths without significant performance degradation. Conclusion: We developed a practical and scalable system for automated PVD annotation that performed well across all subtypes. This work provides a reproducible foundation for automated PVD labeling to support the development of machine-learning-based clinical decision support systems for identifying patient-level asynchrony.

Rheumatic heart disease (RHD) remains a major public health concern across low- and middle-income countries in the Global South. Early detection through community-based screening of asymptomatic individuals has been identified as a critical strategy for reducing the disease burden. Despite this, the absence of accessible, automated population screening tools continues to impede implementation at scale. This study investigates the screening potential of integrating electrocardiography (ECG) and phonocardiography (PCG) for the early detection of RHD in asymptomatic schoolchildren. The dataset was obtained as part of an ambulatory screening initiative conducted across multiple school sites in rural areas of Ethiopia. It comprised ECG and PCG recordings from 611 asymptomatic schoolchildren aged 10 to 20 years. A comprehensive set of time-frequency, visibility graph and non-linear features were extracted from both signal modalities. These features were subsequently evaluated using machine learning models to assess their utility in the automated screening of early RHD. The best model achieved an average 10-folds cross-validation scores on sensitivity, positive-predictive-value and F1-score of 59.6%, 63.6% and 60.8%, respectively for multimodal ECG and PCG signals. Whereas separate evaluation of ECG showed an F1-score of 61.1% and PCG achieved 23.5%. Key features included the T-wave, the area under the QRS complex, and entropy measures derived from beat visibility graphs in the ECG. In addition, visibility graph features from multi-band S1 and S2 heart sound segments, along with MFCC coefficients from the PCG, were also relevant. However, PCG alone performed poorly and did not show improved results over the ECG features. Although auscultation is key clinical diagnosis tool in symptomatic RHD, combined PCG with ECG features does not enhance asymptomatic RHD detection using the ECG modality alone.

Background: Apnoea of prematurity is common and may cause desaturation and/or bradycardia. There is marked variability in infants cardiorespiratory responses to apnoea, despite standardised clinical thresholds. Factors influencing apnoea-related cardiorespiratory instability and whether instability can be predicted warrant investigation. Methods: 181,511 apnoeas >5 seconds were identified from continuous physiological recordings from 146 preterm infants <37 weeks postmenstrual age. Cardiorespiratory instability was defined as bradycardia (>30% heart rate reduction) and/or oxygen desaturation (<85%). Mixed-effects models assessed clinical, demographic and dynamic modulators of the relationship between apnoea duration and cardiorespiratory instability. Machine learning (XGBoost) was used to train models to predict apnoea-related cardiorespiratory instability. Results: Longer duration apnoeas were associated with increased instability, although variability was substantial and 3.6% of apnoeas <10 seconds were associated with cardiorespiratory instability, while 61.2% of apnoeas [&ge;]20 seconds were not. Multiple clinical/demographic (postmenstrual and gestational age, sex, weight z-score, and ventilation mode) and dynamic (baseline heart rate, oxygen saturation, and recent apnoea clustering) factors were associated with increased instability risk. Apnoea-related cardiorespiratory instability could be predicted with a balanced test accuracy of 75.8% when incorporating all features, while a model using only clinical/demographic features achieved 66.0%. Conclusions: Multiple factors influence cardiorespiratory responses to apnoea. Predictive modelling may enable personalised apnoea definitions, improving individualised care.

Background Ventilator-associated pneumonia (VAP) is the most frequent nosocomial infection in critical care, affecting 20-36% of mechanically ventilated patients. Early prediction is hampered by the absence of a reliable, objective diagnostic standard. We developed ADVISE (Automated Dudley Ventilation Infection Series Evaluation), a machine learning model to predict physiological deterioration consistent with developing VAP using routinely collected electronic health record data from a UK NHS intensive care unit. Methods Retrospective observational study of admissions at Russell's Hall Hospital ICU (2008-2026). Following National Data Opt-Out exclusion (158 admissions, 4.2%), 3,566 admissions generated 33,208 candidate 48-hour observation blocks. Six temporal variables - FiO2, ventilator mode, P:F ratio, procalcitonin (PCT), secretion amount, and secretion description - were extracted across the baseline window (hours 1-24). A composite VAP-surrogate outcome required concurrent P:F ratio decline (>=5%) and PCT rise (>=0.5 ng/mL) across the outcome window (hours 25-48). After sequential quality filters, 2,134 blocks (18 positive, 0.84% prevalence) were retained. An XGBoost classifier was trained using nested 5-fold cross-validation with scale_pos_weight=114.0 and ROC-based hyperparameter optimisation on 1,495 training blocks, evaluated on 639 held-out test blocks. Performance was assessed via AUROC, AUPRC, and calibration (Brier score). Bootstrap resampling (1,000 iterations) generated 95% confidence intervals. Results On the held-out test set (n=639, 5 positive outcomes), ADVISE achieved AUROC 0.874 [95% CI: 0.771-0.939] and AUPRC 0.031 [0.008-0.069], representing a 4.0-fold improvement over the no-skill baseline. Nested cross-validation mean AUROC was 0.844 +/- 0.078 (range 0.716-0.915). At the Youden-optimal threshold, sensitivity was 0% with specificity 97.8%, reflecting extreme class imbalance (0.78% test prevalence). A threshold targeting 80% sensitivity achieved sensitivity 80.0% [33.3-100.0%], specificity 87.4% [84.8-89.9%], positive predictive value 4.8% [1.1-9.9%], and negative predictive value 99.8% [99.4-100.0%], detecting 4 of 5 VAP cases with approximately 80 false alarms (12.6% false positive rate). Brier score was 0.0078. Feature importance identified baseline P:F ratio as the dominant predictor (41.3% total gain), followed by ventilator mode (26.1%), secretion amount (13.2%), secretion description (9.1%), procalcitonin (5.9%), and FiO2; (4.5%). Conclusions ADVISE demonstrates that baseline oxygenation trajectory and ventilatory support patterns - derived exclusively from routinely charted ICCA variables - can identify admissions at risk of VAP-related physiological deterioration with meaningful discrimination (AUROC 0.874) despite severe class imbalance. The 80% sensitivity operating point offers a clinically actionable alert rate (12.6% FPR), supporting integration into existing ICU workflows. This proof-of-concept study establishes feasibility; multi-site prospective validation is required before clinical deployment.

The endotracheal tube resistance dominates the total airway resistance in most intubated patients. Mucus deposition and biofilm formation can rapidly increase tube resistance and thereby contribute to serious ventilatory impairments, including dynamic hyperinflation, intrinsic PEEP build-up, added work of breathing, and patient-ventilator asynchrony. During controlled mechanical ventilation, an increased tube resistance can be inferred from the difference between peak and plateau pressure, but this approach fails during pressure-supported spontaneous breathing. Here, we present a method that estimates the linear and nonlinear components of tube resistance from naturally occurring airway pressure and flow fluctuations at the airway opening, without a tracheal pressure sensor and without applying mandatory forced oscillations. This is achieved by solving the equation of motion using band-pass filtered airway pressure and flow signals. Band-pass filtering isolates the relevant resistive and inertive pressure losses across the tube by removing slow contributions from muscle pressure and lung elastance as well as high-frequency noise. The method accurately recovers both linear and nonlinear tube resistance parameters with < 10% error and < 2% bias. Moreover, it enables real-time implementation of full Automatic Tube Compensation (ATC), even in the presence of severe tube obstructions. Continuous estimation of endotracheal tube resistance from naturally occurring airway pressure and flow fluctuations enables real-time detection of clinically relevant tube narrowing and may help improve patient safety, reduce patient-ventilator asynchrony, and facilitate weaning.

Blood pressure (BP) is a vital sign which is measured to diagnose and manage hypertension. However, current methods to measure BP use inflatable cuffs which cause discomfort and limit the frequency at which measurements can be made, or intra-arterial catheters which are invasive and pose infection risks. Here, we propose and evaluate the use of Diffuse Speckle Pulsatile Flowmetry (DSPF) as a cuffless BP measurement method to address these limitations. DSPF is a laser speckle-based technique which simultaneously records blood flow rate and blood volume (i.e. photoplethysmography or PPG) signals from relatively deep vascular tissue. Using information from these signals, we studied DSPFs effectiveness in measuring systolic BP (SBP) and diastolic BP (DBP) through an outpatient study in which 133 patients were recruited, and in measuring beat-to-beat BP waveforms through an inpatient study in which two patients were recruited. In the outpatient study, the DSPF method was able to achieve mean absolute errors (MAEs) of 4.17 mmHg and 2.42 mmHg for SBP and DBP respectively compared to conventional cuff-based methods. It was also able to fulfil the requirements of the AAMI/ESH/ISO 81060-2:2018 standard for BP measurement devices and attain an "A" grade according to the British Hypertension Society grading scheme. For the inpatient study, it produced BP waveforms which had MAEs of 2.35 mmHg and 3.06 mmHg compared to arterial-line measurements for the two patients, respectively. Compared to PPG which has been studied more extensively as a cuffless BP measurement method, we found through ablation studies that DSPF was able to reach significantly lower MAEs and hence better accuracies. DSPF augments the performance of PPG-only methods by leveraging additional information from the blood flow rate signal, and we therefore find it to be a superior cuffless BP measurement method which can potentially be used in outpatient, inpatient, and remote settings.

Cardiovascular disease is the leading cause of death worldwide. Sudden cardiac death (SCD) accounts for roughly 50% of all cardiac deaths. The electrocardiogram (ECG) is widely used for early diagnosis of cardiac disease. However, the complexity of accurate interpretation limits the ECG's efficacy. Modern deep learning methods have been applied to assist clinicians in diagnosis. We applied Neural Architecture Search (NAS), an automated machine learning technique, to identify optimal deep learning architectures for classifying cardiac arrhythmias from ECGs. We applied the Differentiable Architecture Search strategy to an AutoFormer search space to identify optimal self-attention architectures for arrhythmia classification. We trained, validated, and tested the resulting model on the PhysioNet Challenge 2021 dataset (n = 88,253), comprising ECGs across three continents. We performed a hyperparameter optimisation on the NAS output, exploring input patch size, class weighting, and loss function. We evaluated performance using the PhysioNet Challenge metric and the area under the receiver operating characteristic curve (AUROC). The NAS converged towards minimal architectural configurations (embedding dimension: 384, depth: 4, self-attention heads: 4, MLP ratio: 1) with a validation challenge metric of 0.66 (PhysioNet Challenge 21 Winner: 0.63). The NAS-created network achieved an AUROC of 0.97 and a challenge metric of 0.71 during testing. Normal Sinus Rhythm and Sinus Tachycardia achieved AUROCs of 0.99. Low-QRS Voltage and T-wave abnormality were the worst-performing arrhythmias, with AUROCs of 0.89 and 0.90, respectively. We interpret that architectural simplicity drives performance in arrhythmia classification. Because SCD is unexpected, prevention strategies in free-living environments require lightweight computational resources suitable for wearable devices. Class imbalance fundamentally limits classification performance for rare arrhythmias such as Low-QRS Voltage and T-wave inversion, irrespective of hyperparameter choices. However, the self-attention mechanism can autonomously abstract clinical representations, simplifying clinical deployment by eliminating the need for an explicit feature-extraction pipeline.

Background and Aims: Clinical interpretation of the precordial leads V1-V6 assumes that Wilson's central terminal (WCT) has a fixed anatomical location. Consequently, a positive signal corresponds to electrical activation spreading from WCT towards the respective electrode, and vice versa. However, the location of WCT has never been systematically investigated. Yet, a better understanding of WCT location could improve the interpretation of the precordial leads. This work aims to characterize the spatial expansion and location of the physical WCT i.e., the electrical potential defined by the WCT, during the P-wave on the body surface. Methods: An intensive analysis of body surface potential maps (BSPMs) during atrial depolarization in an in silico patient cohort and clinical data was conducted. Results: During the P-wave, the location of WCT was not stationary but the spatial extent and location varied across time as well as across individuals. Four distinct spatial patterns of WCT distribution on the body surface were identified in silico, and three of these were found in the clinical cohort. WCT signals agreed with BSPM signals at commonly assumed positions of WCT only for a small fraction of the P-wave. Conclusion: The spatial extension and location of WCT changes during the P-wave and thus should be considered when interpreting the precordial leads.

Body surface gastric mapping (BSGM) enables non-invasive measurement of gastric electrophysiology, but the signals are approximately 100 times weaker than cardiac potentials and overlap spectrally with motion artifacts, necessitating labor-intensive manual review that limits clinical scalability. We present an uncertainty-aware deep learning framework combining a signal reconstruction network with a parallel uncertainty estimation network to automate artifact correction in high-resolution BSGM. Models were trained on 2,398 multihour, 64-channel recordings from 27 international clinical sites using weak supervision, a physiology-aware loss function, and uncertainty-gated quality control. In an independent cohort of 127 patients, the system achieved relative reductions of 39% in signal reconstruction error, 9% in total data removed, and 23% in amplitude--movement correlation compared with the industry-standard Wiener filter. Improved signal fidelity altered automated clinical phenotyping in 7% of patients by recovering previously obscured gastric rhythms. Uncertainty-aware deep learning enables reliable automated artifact correction in body-surface gastric mapping, improving signal fidelity and enabling scalable clinical interpretation. The system is FDA-cleared (510(k) K252504) and deployed in clinical practice, demonstrating that data-driven artifact correction can meet regulatory requirements for medical devices and reduce dependence on specialist manual review.

A Wearable Monitoring System (WMS), comprising a chest patch, wrist-worn pulse oximeter, and arm-worn blood pressure device, was developed in preparation for a pilot Randomised Controlled Trial (RCT) on a UK surgical ward. The system was designed to support continuous physiological monitoring and early detection of deterioration. An initial prototype user interface was developed by the research team based on prior clinical experience and engineering knowledge. To ensure suitability for clinical practice, iterative user-centred refinement was undertaken through a series of clinician focus groups and wearability assessments. Six focus groups were conducted between November 2019 and May 2021 involving multidisciplinary healthcare professionals. Feedback from these sessions informed successive interface and system modifications. System development spanned the COVID-19 pandemic, during which the WMS was rapidly adapted and deployed to support clinical care on isolation wards. Feedback obtained during this period was incorporated into later versions of the system and provided a unique opportunity to examine changes in clinician priorities under pandemic conditions. Clinicians consistently prioritised alert visibility, alarm fatigue mitigation, parameter flexibility, and centralised monitoring. Notably, preferences regarding alert modality and access mechanisms evolved over time: early enthusiasm for mobile or smartphone-type devices shifted towards a preference for fixed, ward-based displays and audible alerts at the nurses station following pandemic deployment. Building on previous wearability testing in healthy volunteers, wearability testing using a validated questionnaire was completed by 169 patient participants during the RCT. The chest patch and pulse oximeter demonstrated high tolerability, whereas the blood pressure cuff showed poor wearability and was removed from the final system. These findings demonstrate the importance of iterative, clinician-led design for wearable WMS and highlight how extreme clinical contexts such as the COVID-19 pandemic can significantly reshape perceived requirements for safety-critical monitoring technologies.