Multi‐parametric system for risk stratification in mitral regurgitation: A multi‐task Gaussian prediction approach

Tse, Gary; Zhou, Jiandong; Lee, Sharen; Liu, Yingzhi; Leung, Keith Sai Kit; Lai, Rachel Wing Chuen; Burtman, Anthony David; Wilson, Carly; Liu, Tong; Li, Ka Hou Christien; Lakhani, Ishan; Zhang, Qingpeng

doi:10.1111/eci.13321

Cited by 15 publications

(14 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification of HF into those with reduced, midrange, and preserved ejection fraction is important, 7 and risk stratification strategies depend partly on the behaviour of LVEF 8 , 9 and require a multi‐parametric approach. 10 , 11 , 12 , 13 LVEF may recover, remain stable, or decline owing to a complex interaction between comorbidities, frailty status, 14 , 15 , 16 , 17 , 18 , 19 medical or device treatment, 20 and disease progression. Our results are consistent with the results of other studies; compared with persistent HFrEF, subjects with HFrecEF are younger, with a higher prevalence of hypertension, lower prevalence of coronary artery disease, cerebrovascular disease, and diabetes mellitus, 3 , 21 and had higher interventricular septal thickness, lower left ventricular diameter, and a better biomarker indicator.…”

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Characteristics and outcomes of heart failure with recovered left ventricular ejection fraction

et al. 2021

Self Cite

View full text Add to dashboard Cite

Aims There is an emerging interest in elucidating the natural history and prognosis for patients with heart failure with reduced ejection fraction (HFrEF) in whom left ventricular ejection fraction (LVEF) subsequently improves. The characteristics and outcomes were compared between heart failure with recovered ejection fraction (HFrecEF) and persistent HFrEF. Methods and results This is a retrospective study of adults who underwent at least two echocardiograms 3 months apart between 1 November 2015 and 31 October 2019 with an initial diagnosis of HFrEF. The subjects were divided into HFrecEF group (second LVEF > 40%, ≥10% absolute improvement in LVEF) and persistent HFrEF group (<10% absolute improvement in LVEF) according to the second LVEF. To further study the characteristics of HFrecEF patients, the cohort was further divided into LVEF improvement of 10–20% and >20% subgroups. The primary outcomes were all‐cause mortality and rehospitalization. A total of 1160 HFrEF patients were included [70.2% male, mean (standard deviation) age: 62 ± 13 years]. On the second echocardiogram, 284 patients (24.5%) showed HFrecEF and 876 patients (75.5%) showed persistent HFrEF. All‐cause mortality was identified in 23 (8.10%) HFrecEF and 165 (18.84%) persistent HFrEF, whilst 76 (26.76%) and 426 (48.63%) showed rehospitalizations, respectively. Survival analysis showed that the persistent HFrEF subgroup experienced a significantly higher mortality at 12 and 24 months and a higher hospitalization at 12, 24, 48, and more than 48 months following discharge. Multivariate Cox regression showed that persistent HFrEF had a higher risk of all‐cause mortality [hazard ratio (HR) 2.30, 95% confidence interval (CI) 1.49–3.56, P = 0.000] and rehospitalization (HR 1.85, 95% CI 1.45–2.36, P = 0.000) than the HFrecEF group. Subgroup analysis showed that the LVEF ≥ 20% improvement subgroup had lower rates of adverse outcomes compared with those with less improvement of 10–20%. Conclusions Heart failure with recovered ejection fraction is a distinct HF phenotype with better clinical outcomes compared with those with persistent HFrEF. HFrecEF patients have a relatively better short‐term mortality at 24 months but not thereafter.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Characteristics and outcomes of heart failure with recovered left ventricular ejection fraction

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…35 In this study, NLR was found to predict all-cause mortality in our cohort of HF patients with LVEF ≤ 45%. NLR can also predict adverse outcomes in other cardiac diseases such as mitral regurgitation 36 and aortic stenosis. 37 Moreover, poor nutrition leads to impaired immune responses and leads to adverse outcomes in HF.…”

Section: Inflammatory and Nutrition Indicesmentioning

confidence: 99%

Multi‐modality machine learning approach for risk stratification in heart failure with left ventricular ejection fraction ≤ 45%

Tse

Zhou

Woo³

et al. 2020

ESC Heart Failure

Self Cite

View full text Add to dashboard Cite

Aims Heart failure (HF) involves complex remodelling leading to electrical and mechanical dysfunction. We hypothesized that machine learning approaches incorporating data obtained from different investigative modalities including atrial and ventricular measurements from electrocardiography and echocardiography, blood inflammatory marker [neutrophil-to-lymphocyte ratio (NLR)], and prognostic nutritional index (PNI) will improve risk stratification for adverse outcomes in HF compared to logistic regression. Methods and results Consecutive Chinese patients referred to our centre for transthoracic echocardiography and subsequently diagnosed with HF, between 1 January 2010 and 31 December 2016, were included in this study. Two machine learning techniques, multilayer perceptron and multi-task learning, were compared with logistic regression for their ability to predict incident atrial fibrillation (AF), transient ischaemic attack (TIA)/stroke, and all-cause mortality. This study included 312 HF patients [mean age: 64 (55-73) years, 75% male]. There were 76 cases of new-onset AF, 62 cases of incident TIA/stroke, and 117 deaths during follow-up. Univariate analysis revealed that age, left atrial reservoir strain (LARS) and contractile strain (LACS) were significant predictors of new-onset AF. Age and smoking predicted incident stroke. Age, hypertension, type 2 diabetes mellitus, chronic kidney disease, mitral or aortic regurgitation, P-wave terminal force in V1, the presence of partial inter-atrial block, left atrial diameter, ejection fraction, global longitudinal strain, serum creatinine and albumin, high NLR, low PNI, and LARS and LACS predicted all-cause mortality. Machine learning techniques achieved better prediction performance than logistic regression. Conclusions Multi-modality assessment is important for risk stratification in HF. A machine learning approach provides additional value for improving outcome prediction.

show abstract

“…Currently, there is yet a multi-parametric approach in the risk stratification of MR. Recently, we reported that risk stratification of MR can be significant improved with the use of a multi-task Gaussian process learning model which outperformed logistic regression [17]. In this study, we extend previous analyses by assessing the comparative performance of several machine learning models, such as Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Network (ANN), and Gradient Boosting Machine (GBM).…”

Section: Introductionmentioning

confidence: 76%

Predicting Stroke and Mortality in Mitral Regurgitation: A Gradient Boosting Approach

Zhou

Lee²,

Liu³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

IntroductionWe hypothesized that an interpretable gradient boosting machine (GBM) model considering comorbidities, P-wave and echocardiographic measurements, can better predict mortality and cerebrovascular events in mitral regurgitation (MR).MethodsPatients from a tertiary center were analyzed. The GBM model was used as an interpretable statistical approach to identify the leading indicators of high-risk patients with either outcome of CVAs and all-cause mortality.ResultsA total of 706 patients were included. GBM analysis showed that age, systolic blood pressure, diastolic blood pressure, plasma albumin levels, mean P-wave duration (PWD), MR regurgitant volume, left ventricular ejection fraction (LVEF), left atrial dimension at end-systole (LADs), velocity-time integral (VTI) and effective regurgitant orifice were significant predictors of TIA/stroke. Age, sodium, urea and albumin levels, platelet count, mean PWD, LVEF, LADs, left ventricular dimension at end systole (LVDs) and VTI were significant predictors of all-cause mortality. The GBM demonstrates the best predictive performance in terms of precision, sensitivity c-statistic and F1-score compared to logistic regression, decision tree, random forest, support vector machine, and artificial neural networks.ConclusionGradient boosting model incorporating clinical data from different investigative modalities significantly improves risk prediction performance and identify key indicators for outcome prediction in MR.

show abstract

Multi‐parametric system for risk stratification in mitral regurgitation: A multi‐task Gaussian prediction approach

Cited by 15 publications

References 40 publications

Characteristics and outcomes of heart failure with recovered left ventricular ejection fraction

Characteristics and outcomes of heart failure with recovered left ventricular ejection fraction

Multi‐modality machine learning approach for risk stratification in heart failure with left ventricular ejection fraction ≤ 45%

Predicting Stroke and Mortality in Mitral Regurgitation: A Gradient Boosting Approach

Contact Info

Product

Resources

About