Publications by authors named "Linyuan Jing"
Article Synopsis
- A novel deep learning (DL) system was developed to enhance the interpretation of transthoracic echocardiography (TTE) for assessing the severity of mitral regurgitation (MR) by integrating multiple video assessments.
- The system was tested with a large dataset (over 61,000 TTEs) and showed high accuracy in classifying MR severity, achieving exact accuracy rates of 82% for internal and 79% for external test sets.
- Most misclassifications occurred between none/trace and mild MR categories, and the use of multiple TTE views improved classification accuracy.
StrainNet: Improved Myocardial Strain Analysis of Cine MRI by Deep Learning from DENSE.
Radiol Cardiothorac Imaging
June 2023
Article Synopsis
- A study developed a deep learning model called StrainNet that analyzes heart displacement and strain using cine MRI data and DENSE measurements.
- It involved training and testing the model on data gathered from a diverse group of patients with heart diseases and healthy individuals over several years, focusing on the model's accuracy in predicting myocardial movements.
- The results indicated that StrainNet performed better than traditional feature tracking methods, showing strong agreement with DENSE measurements for both global and segmental strain analysis.
Background: Several large trials have employed age or clinical features to select patients for atrial fibrillation (AF) screening to reduce strokes. We hypothesized that a machine learning (ML) model trained to predict AF risk from 12‑lead electrocardiogram (ECG) would be more efficient than criteria based on clinical variables in indicating a population for AF screening to potentially prevent AF-related stroke.
Methods: We retrospectively included all patients with clinical encounters in Geisinger without a prior history of AF.
Article Synopsis
- The study focuses on improving the diagnosis of structural heart diseases through a new ECG-based machine learning model that predicts various conditions, potentially increasing patient outcomes.
- By analyzing 2.2 million ECGs linked to health records, researchers tested their model on seven echocardiography-confirmed diseases, ultimately achieving a high predictive accuracy (0.91 AUC) with a 42% positive predictive value.
- The composite model outperformed individual disease predictions and showed consistent results across different datasets, emphasizing the effectiveness of incorporating diverse patient data for better heart disease detection.
Article Synopsis
- Atrial fibrillation (AF) can lead to serious health issues if not detected early, and the study aims to use a deep neural network to predict new-onset AF from resting 12-lead ECGs in patients without a previous AF history.
- Researchers analyzed 1.6 million ECG traces from 430,000 patients, achieving good predictive performance with an area under the receiver operating characteristic curve of 0.85, indicating an effective ability to identify those at risk for AF within a year.
- The model also demonstrated that it could indicate a high risk for AF-related strokes, successfully identifying 62% of patients who experienced such strokes within three years, thereby highlighting the potential for targeted screening strategies.
Machine learning promises to assist physicians with predictions of mortality and of other future clinical events by learning complex patterns from historical data, such as longitudinal electronic health records. Here we show that a convolutional neural network trained on raw pixel data in 812,278 echocardiographic videos from 34,362 individuals provides superior predictions of one-year all-cause mortality. The model's predictions outperformed the widely used pooled cohort equations, the Seattle Heart Failure score (measured in an independent dataset of 2,404 patients with heart failure who underwent 3,384 echocardiograms), and a machine learning model involving 58 human-derived variables from echocardiograms and 100 clinical variables derived from electronic health records.
View Article and Find Full Text PDFArticle Synopsis
- The study investigates using a deep neural network (DNN) to predict 1-year all-cause mortality from electrocardiogram (ECG) voltage-time data collected over 34 years.
- The DNN was trained on over 1.1 million ECGs from nearly 253,400 patients, achieving a high accuracy (AUC of 0.88) in predicting mortality, even among patients whose ECGs were deemed 'normal' by doctors.
- Results indicate that the DNN can uncover significant prognostic insights that may not be apparent to cardiologists, with a notable hazard ratio of 9.5 for predicting 1-year mortality.
Article Synopsis
- Heart failure is a common and expensive condition, prompting the need for better management strategies that utilize population health data and machine learning.
- By analyzing electronic health record data from Geisinger, the study aimed to predict 1-year mortality for patients with heart failure by examining various clinical metrics and care gaps.
- The study found that a machine learning model (XGBoost) effectively identified 2,844 patients at high risk of death, suggesting that addressing identified care gaps could potentially save 231 lives within a year.
Article Synopsis
- Arrhythmogenic right ventricular cardiomyopathy (ARVC) is linked to specific gene variants, but their occurrence and effects in the general population remain unclear.
- Researchers analyzed data from 61,019 individuals to find loss-of-function (LOF) variants related to ARVC and assessed their clinical significance through ECG and echocardiogram evaluations.
- The study found ARVC variants in 0.23% of individuals, but these individuals generally showed no significant heart-related symptoms, indicating a low clinical impact and underscoring the need for further detailed studies.
Article Synopsis
- The study examined the link between left ventricular ejection fraction (LVEF) and survival rates in a large group of patients over 20 years, using a multitude of echocardiogram data.
- Findings revealed a u-shaped relationship between LVEF and mortality, with the lowest risk of death at LVEF levels of 60-65%, and increased risks for values above 70% and below 35-40%.
- The results suggest that having an LVEF outside the 60-65% range is linked to worse survival outcomes, indicating the potential identification of a new heart function phenotype associated with high LVEF.
Comparison of left ventricular strains and torsion derived from feature tracking and DENSE CMR.
J Cardiovasc Magn Reson
September 2018
Article Synopsis
- Cardiovascular magnetic resonance (CMR) feature tracking is gaining popularity for measuring heart mechanics, but its accuracy compared to established techniques like DENSE imaging is still uncertain.
- A study with 88 participants revealed that while feature tracking showed acceptable agreement for measuring mid-ventricular circumferential strain, it overestimated values at the base and apex of the heart and underestimated torsion.
- The findings suggest that while feature tracking can be useful, its measurements can vary significantly, indicating the need for careful interpretation in clinical settings.
Article Synopsis
- The study aimed to enhance survival predictions after echocardiography using machine learning techniques rather than just traditional methods relying on ejection fraction and comorbidities.
- Researchers analyzed data from over 171,000 patients and compared the effectiveness of nonlinear machine learning models against standard logistic regression models using multiple data inputs.
- Results showed that machine learning models, particularly random forest models incorporating extensive echocardiographic data, significantly outperformed traditional clinical risk assessments in predicting patient survival outcomes.
Article Synopsis
- Patients with Ebstein anomaly are at risk for issues like left ventricular dysfunction and dyssynchrony, in addition to tricuspid regurgitation and right ventricular enlargement, making surgical intervention important.
- A study on cone tricuspid valve reconstruction showed significant improvements post-surgery, including a reduction in tricuspid regurgitation and right ventricular volume, alongside increased left ventricular size and stroke volume, though ejection fractions remained unchanged.
- The operation also enhanced the synchrony of left ventricular contractions, as indicated by a decreased dyssynchrony index, which is related to the improvements in right ventricular volume and tricuspid regurgitation.
Article Synopsis
- - This study aimed to determine which patients with repaired tetralogy of Fallot (rTOF) are at risk for worsening heart conditions by employing a machine learning approach, rather than relying on traditional regression methods.
- - Researchers analyzed clinical data from 153 rTOF patients using cardiac magnetic resonance (CMR) imaging over time, categorizing the deterioration into three levels: none, minor, or major, and identified key predictors for deterioration using a support vector machine classifier.
- - The findings highlighted that the machine learning models effectively predicted deterioration in ventricular function, with crucial baseline indicators like left ventricular ejection fraction, circumferential strain, and pulmonary regurgitation, enabling potential early interventions for at-risk patients.
Association between left ventricular mechanics and diffuse myocardial fibrosis in patients with repaired Tetralogy of Fallot: a cross-sectional study.
Christopher M Haggerty Jonathan D Suever Arichanah Pulenthiran Abba Mejia-Spiegeler Gregory J Wehner Linyuan Jing
J Cardiovasc Magn Reson
December 2017
Article Synopsis
- Patients with repaired tetralogy of Fallot (TOF) experience progressive biventricular remodeling, leading to issues with heart function, particularly due to diffuse myocardial fibrosis.
- A study involving 40 TOF patients used cardiac magnetic resonance imaging to investigate the relationship between diffuse fibrosis measures and left ventricular (LV) mechanics, focusing on factors like extracellular volume fraction (ECV).
- Results showed that while some LV mechanics were moderately impaired, there were significant associations between ECV and LV dyssynchrony index as well as peak radial strain, suggesting that diffuse fibrosis may contribute to ventricular dysfunction in TOF patients.
Article Synopsis
- Children with obesity often experience changes in heart structure, known as hypertrophic cardiac remodeling, partly due to high blood pressure (hypertension).
- A study involving 72 children, aged 8-17, found that those who were obese or overweight had significantly higher measures of heart mass and thickness compared to their healthy-weight peers, with 35% showing signs of concentric hypertrophy.
- Blood pressure readings indicated that 26% of the obese/overweight children had ambulatory hypertension, and there were strong correlations between body mass index (BMI) and various heart measurements, suggesting that both obesity and blood pressure contribute to heart changes in children.
Article Synopsis
- Pediatric obesity is a rising public health issue linked to serious health risks like heart disease and early death, highlighting the need to explore right ventricular (RV) changes in addition to left ventricular (LV) issues in obese children.
- A study involving 103 children aged 8-18 used advanced imaging techniques to assess the geometry and function of the RV, discovering significant differences between healthy-weight and obese/overweight groups.
- Results showed that obese/overweight children had a 22% increase in RV mass and poorer RV function (longitudinal strain), with some exhibiting more severe conditions like LV concentric hypertrophy, which further impaired RV functionality.
Article Synopsis
- The mechanics of the right ventricle (RV) are difficult to study due to its thin walls and complex shape, but a new imaging technique called 3D DENSE allows for better analysis of RV function.
- In a study of 50 healthy individuals, researchers compared RV mechanics (like strain, torsion, and synchrony) to those of the left ventricle (LV) using advanced mathematical models to interpret 3D cardiac data.
- Results showed that while global circumferential strain was similar for both ventricles, the RV had greater longitudinal strain and contracted more synchronously than the LV, indicating important differences in their mechanics.
Article Synopsis
- Patients with repaired tetralogy of Fallot (rTOF) experience long-term heart issues, specifically progressive ventricular dysfunction, after surgical repair.
- A study analyzed patients who had cardiovascular magnetic resonance (CMR) scans to assess if measures of heart strain and dyssynchrony could predict worsening heart function over time.
- Results showed that key predictors like ventricular strain and dyssynchrony did not significantly correlate with changes in ventricular size and function, indicating that current prediction methods are limited for rTOF patients.
Article Synopsis
- Obesity affects about 20% of children and is linked to a higher risk of heart disease and early death, prompting a study using cardiovascular magnetic resonance (CMR) to assess heart changes in obese kids.
- The study involved 41 obese/overweight children and 29 healthy children, revealing that obese kids had significantly larger left ventricular (LV) mass and thicker myocardium, with cardiac changes observable as early as age 8.
- Findings indicated that 24% of obese children had concentric hypertrophy, with this group showing the greatest impairment in heart function, suggesting they may be at higher risk and need closer monitoring.
Spatially discordant alternans (DA) of action potential durations (APD) is thought to be more pro-arrhythmic than concordant alternans. Super normal conduction (SNC) has been reported to suppress formation of DA. An increase in conduction velocity (CV) as activation rate increases, i.
View Article and Find Full Text PDFArticle Synopsis
- DENSE, a technique in MRI, encodes tissue displacement in its signal phase using different encoding frequencies (ke), which affect image quality and sensitivity to movement.
- A study tested various ke values (0.02 to 0.10 cycles/mm) in healthy subjects and heart disease patients to evaluate cardiac mechanics and image characteristics.
- Results indicated that using ke of 0.04 cycles/mm provides high accuracy in measuring cardiac mechanics with zero phase wrapping and better signal-to-noise ratio compared to the higher ke values.
Article Synopsis
- Cardiovascular magnetic resonance (CMR) using cine balanced steady state free precession (bSSFP) MRI can differentiate myocardial tissue types based on their signal intensity related to magnetization transfer (MT) and T1/T2-relaxation times.
- A study involving 47 patients and 10 healthy controls tested a method called 2-point bSSFP, which compares images taken with two different flip angles, to detect areas of enhanced tissue similar to gadolinium-enhanced imaging.
- The results indicated that 2-point bSSFP showed good agreement with standard late gadolinium enhancement (LGE) imaging, suggesting it could be a valuable, gadolinium-free alternative for identifying myocardial tissue abnormalities.