A5069641751- ECG Monitoring and Analysis
- EEG and Brain-Computer Interfaces
- Blind Source Separation Techniques
- Phonocardiography and Auscultation Techniques
- Heart Rate Variability and Autonomic Control
- Cardiac electrophysiology and arrhythmias
- Neonatal and fetal brain pathology
- Analog and Mixed-Signal Circuit Design
- Non-Invasive Vital Sign Monitoring
- Speech and Audio Processing
- COVID-19 epidemiological studies
- Neural dynamics and brain function
- Neural Networks and Applications
- Congenital Heart Disease Studies
- Nursing Diagnosis and Documentation
- Healthcare Technology and Patient Monitoring
- Gaze Tracking and Assistive Technology
- Fault Detection and Control Systems
- Spectroscopy and Chemometric Analyses
- Cardiac Arrest and Resuscitation
- Cardiac Arrhythmias and Treatments
- Neuroscience and Neural Engineering
- Machine Learning in Bioinformatics
- Topic Modeling
- Traumatic Brain Injury Research
Emory University
2020-2025
Georgia Institute of Technology
2024-2025
Emory and Henry College
2024
University of California, San Francisco
2023
Massachusetts General Hospital
2023
Shiraz University
2013-2022
University School
2021
Grenoble Images Parole Signal Automatique
2007-2020
Université Grenoble Alpes
2006-2020
Centre National de la Recherche Scientifique
2006-2020
Abstract In the past few decades, analysis of heart sound signals (i.e. phonocardiogram or PCG), especially for automated segmentation and classification, has been widely studied reported to have potential value detect pathology accurately in clinical applications. However, comparative analyses algorithms literature hindered by lack high-quality, rigorously validated, standardized open databases recordings. This paper describes a public database, assembled an international competition,...
In this paper, a nonlinear Bayesian filtering framework is proposed for the of single channel noisy electrocardiogram (ECG) recordings. The necessary dynamic models ECG are based on modified model, previously suggested generation highly realistic synthetic ECG. A version model used in several filters, including Extended Kalman Filter, Smoother, and Unscented Filter. An automatic parameter selection method also introduced, to facilitate adaptation parameters vast variety ECGs. This approach...
Cardiac auscultation is one of the most cost-effective techniques used to detect and identify many heart conditions. Computer-assisted decision systems based on can support physicians in their decisions. Unfortunately, application such clinical trials still minimal since them only aim presence extra or abnormal waves phonocardiogram signal, i.e., a binary ground truth variable (normal vs abnormal) provided. This mainly due lack large publicly available datasets, where more detailed...
In this letter, we propose the application of generalized eigenvalue decomposition for multichannel electrocardiogram (ECG) recordings. The proposed method uses a modified version previously presented measure periodicity and phase-wrapping RR-interval, extracting "most periodic" linear mixtures recorded dataset. It is shown that an improved extension conventional source separation techniques, specifically customized ECG signals. therefore special interest compression ECG, removal maternal...
A three-dimensional dynamic model of the electrical activity heart is presented. The based on single dipole and later related to body surface potentials through a linear which accounts for temporal movements rotations cardiac dipole, together with realistic ECG noise model. proposed also generalized maternal fetal mixtures recorded from abdomen pregnant women in multiple pregnancies. applicability evaluation signal processing algorithms illustrated using independent component analysis....
In this research, we study the propagation patterns of epidemic diseases such as COVID-19 coronavirus, from a mathematical modeling perspective. The is based on an extensions well-known susceptible-infected-recovered (SIR) family compartmental models. It shown how social measures distancing, regional lockdowns, quarantine and global public health vigilance, influence model parameters, which can eventually change mortality rates active contaminated cases over time, in real world. As with all...
Abstract Cardiac auscultation is an accessible diagnostic screening tool that can help to identify patients with heart murmurs for follow-up and treatment abnormal cardiac function. However, experts are needed interpret the sounds, limiting accessibility of care in resource-constrained environments. Therefore, George B. Moody PhysioNet Challenge 2022 invited teams develop algorithmic approaches detecting function from phonocardiogram (PCG) recordings sounds. For Challenge, we sourced 5272...
Abstract Background Seizure detection is challenging outside the clinical environment due to lack of comfortable, reliable, and practical long-term neurophysiological monitoring devices. We developed a novel, discreet, unobstructive in-ear sensing system that enables electroencephalography (EEG) recording. This first study we are aware systematically compares seizure utility EEG with simultaneously recorded intracranial EEG. In addition, present similar comparison between scalp Methods this...
Abstract Objective. The EPHNOGRAM project aimed to develop a low-cost, low-power device for simultaneous electrocardiogram (ECG) and phonocardiogram (PCG) recording, with additional channels environmental audio enhance PCG through active noise cancellation. objective was study multimodal electro-mechanical activities of the heart, offering insights into differences synergies between these modalities during various cardiac activity levels. Approach. We developed tested several hardware...
Cardiovascular diseases are a major cause of mortality globally, and electrocardiograms (ECGs) crucial for diagnosing them. Traditionally, ECGs stored in printed formats. However, these printouts, even when scanned, incompatible with advanced ECG diagnosis software that require time-series data. Digitizing images is vital training machine learning models diagnosis, leveraging the extensive global archives collected over decades. Deep image processing promising this regard, although lack...
In this paper an extended Kalman filter (EKF) has been proposed for the filtering of noisy ECG signals. The method is based on a modified nonlinear dynamic model, previously introduced generation synthetic An automatic parameter selection also suggested, to adapt model with vast variety normal and abnormal results show that EKF output able track original signal shape even in most noisiest epochs signal. may serve as efficient procedure applications such noninvasive extraction fetal cardiac...
A general deflation framework is described for the separation of a desired signal subspace arbitrary dimensions from noisy multichannel observations. The method simultaneously uses single and priors to split undesired subspaces, even coplanar (intersecting) subspaces. By appropriate use priors, it can extract signals degenerate mixtures noise recorded few number channels in low SNR scenarios, without reduction data dimensions. As case study, performance proposed studied problem extracting...
Frequent and long-term monitoring of fetal health status is still a challenging task in high-risk pregnancies. This paper presents fully noninvasive method to extract heart sound (FHS) from acoustic signals recorded the maternal abdominal surface. The proposed algorithm based on single channel blind source separation (SCBSS), which utilizes empirical mode decomposition (EMD) nonnegative matrix factorization (NMF) extracts different sources audio signal mixtures. To evaluate performance,...
Cardiac wall motion abnormalities (WMA) are strong predictors of mortality, but current screening methods using Q waves from electrocardiograms (ECGs) have limited accuracy and vary across racial ethnic groups. This study aimed to identify novel ECG features deep learning enhance WMA detection, referencing echocardiography as the gold standard. We collected echocardiogram data 35,210 patients in California labeled unstructured language parsing echocardiographic reports. A neural network...
ABSTRACT Background International Classification of Diseases (ICD) codes utilized for congenital heart defect (CHD) case identification in datasets have substantial false‐positive (FP) rates. Incorporating machine learning (ML) algorithms following selection by ICD may improve the accuracy CHD identification, enhancing surveillance efforts. Methods Traditional ML methods were applied to four encounter‐level datasets, 2010–2019, 3334 patients with validated diagnoses and at least one code...
Forecasting the near-exact moments of cardiac phases is crucial for several cardiovascular health applications. For instance, forecasts can enable timing specific stimuli (e.g., image or text presentation in psycholinguistic experiments) to coincide with like systole (cardiac ejection) and diastole filling). This capability could be leveraged enhance amplitude a subject's response, prompt them fight-or-flight scenarios conduct retrospective analysis physiological predictive models. While...
ABSTRACT Background International Classification of Disease (ICD) codes can accurately identify patients with certain congenital heart defects (CHDs). In ICD‐defined CHD data sets, the code for secundum atrial septal defect (ASD) is most common, but it has a low positive predictive value CHD, potentially resulting in drawing erroneous conclusions from such sets. Methods reduced false rates among individuals captured ASD ICD are needed public health surveillance. We propose two‐level...