Searcharxiv⌕ Search

arXiv subjects

Aruna Mohan

Publications and source records attributed to Aruna Mohan.

6 recordsLinked to original sources
arXiv2026

A Vision Transformer for ECG-Based Detection of Left Ventricular Systolic Dysfunction Across Multiple Clinical Sites

Reduced left ventricular ejection fraction (LVEF) is frequently asymptomatic and often detected only after advanced heart failure develops. Electrocardiograms are recorded routinely yet underused for this condition, because reduced LVEF has no single diagnostic waveform. We trained an ensemble of vision transformers from scratch to detect reduced LVEF ($\leq$40%) from 12-lead ECGs, analyzing each heartbeat individually, using 10,142 patients across seven sites in three US health systems. In a held-out external cohort of 4,092 patients from three geographically independent US clinical sites at a real-world reduced-LVEF prevalence of 8.72%, the model achieved an AUROC of 0.88 (95% CI 0.86-0.89), sensitivity 81.2%, specificity 81.0%, and negative predictive value 97.8%. Sensitivity remained high across sex, race, ethnicity, and comorbidity subgroups, while specificity was lower in older patients and those with atrial fibrillation or cardiomyopathy. Beat-level attention maps provided interpretability into the model's predictions, showing consistent focus on the QRS complex rather than the P wave. These findings support the potential of routine ECGs as a scalable first-pass triage step to identify patients who should undergo echocardiography for reduced ejection fraction across diverse patient populations.

cs.CV↗
arXiv2026

RhythmBERT: A Self-Supervised Language Model Based on Latent Representations of ECG Waveforms for Heart Disease Detection

Electrocardiogram (ECG) analysis is crucial for diagnosing heart disease, but most self-supervised learning methods treat ECG as a generic time series, overlooking physiologic semantics and rhythm-level structure. Existing contrastive methods utilize augmentations that distort morphology, whereas generative approaches employ fixed-window segmentation, which misaligns cardiac cycles. To address these limitations, we propose RhythmBERT, a generative ECG language model that considers ECG as a language paradigm by encoding P, QRS, and T segments into symbolic tokens via autoencoder-based latent representations. These discrete tokens capture rhythm semantics, while complementary continuous embeddings retain fine-grained morphology, enabling a unified view of waveform structure and rhythm. RhythmBERT is pretrained on approximately 800,000 unlabeled ECG recordings with a masked prediction objective, allowing it to learn contextual representations in a label-efficient manner. Evaluations show that despite using only a single lead, RhythmBERT achieves comparable or superior performance to strong 12-lead baselines. This generalization extends from prevalent conditions such as atrial fibrillation to clinically challenging cases such as subtle ST-T abnormalities and myocardial infarction. Our results suggest that considering ECG as structured language offers a scalable and physiologically aligned pathway for advancing cardiac analysis.

cs.LG↗
arXiv2024

Deciphering Heartbeat Signatures: A Vision Transformer Approach to Explainable Atrial Fibrillation Detection from ECG Signals

Remote patient monitoring based on wearable single-lead electrocardiogram (ECG) devices has significant potential for enabling the early detection of heart disease, especially in combination with artificial intelligence (AI) approaches for automated heart disease detection. There have been prior studies applying AI approaches based on deep learning for heart disease detection. However, these models are yet to be widely accepted as a reliable aid for clinical diagnostics, in part due to the current black-box perception surrounding many AI algorithms. In particular, there is a need to identify the key features of the ECG signal that contribute toward making an accurate diagnosis, thereby enhancing the interpretability of the model. In the present study, we develop a vision transformer approach to identify atrial fibrillation based on single-lead ECG data. A residual network (ResNet) approach is also developed for comparison with the vision transformer approach. These models are applied to the Chapman-Shaoxing dataset to classify atrial fibrillation, as well as another common arrhythmia, sinus bradycardia, and normal sinus rhythm heartbeats. The models enable the identification of the key regions of the heartbeat that determine the resulting classification, and highlight the importance of P-waves and T-waves, as well as heartbeat duration and signal amplitude, in distinguishing normal sinus rhythm from atrial fibrillation and sinus bradycardia.

eess.SP↗
arXiv2008

Polymer translocation through pores with complex geometries

We propose a method for the theoretical investigation of polymer translocation through composite pore structures possessing arbitrarily specified geometries. Translocation through each constituent part of the composite is treated as being analogous to the diffusion of the translocation coordinate over the free energy landscape derived from the chain configurations within the pore. The proposed method accounts for possible reverse motions of the leading chain end at the interface between constituent parts of a composite pore, a possibility that has been neglected in prior studies. As an illustration of our method, we study the translocation of a Gaussian chain between two spherical compartments connected by a cylindrical pore, and by a composite pore consisting of two connected cylinders of different diameters, which is structurally similar to the $α$-hemolysin membrane channel. We demonstrate that reverse chain motions between the pore constituents may contribute significantly to the total translocation time. Our results further establish that translocation through a two-cylinder composite pore is faster when the chain is introduced into the pore on the cis (wide) side of the channel rather than the trans (narrow) side.

cond-mat.soft↗
arXiv2007

Effect of charge distribution on the translocation of an inhomogeneously charged polymer through a nanopore

We investigate the voltage-driven translocation of an inhomogeneously charged polymer through a nanopore by utilizing discrete and continuous stochastic models. As a simplified illustration of the effect of charge distribution on translocation, we consider the translocation of a polymer with a single charged site in the presence and absence of interactions between the charge and the pore. We find that the position of the charge that minimizes the translocation time in the absence of pore--polymer interactions is determined by the entropic cost of translocation, with the optimum charge position being at the midpoint of the chain for a rodlike polymer and close to the leading chain end for an ideal chain. The presence of attractive or repulsive pore--charge interactions yields a shift in the optimum charge position towards the trailing end and the leading end of the chain, respectively. Moreover, our results show that strong attractive or repulsive interactions between the charge and the pore lengthen the translocation time relative to translocation through an inert pore. We generalize our results to accommodate the presence of multiple charged sites on the polymer. Our results provide insight into the effect of charge inhomogeneity on protein translocation through biological membranes.

cond-mat.soft↗