Left atrial overload detection in ECG using frequency domain features with convolutional neural network

Background/Introduction

Volume and pressure alterations in the left atrium (LA) are utilized as important biomarkers due to their association with numerous cardiac pathologies. Because of the anatomical peculiarities of the LA, echocardiographic evaluation is quite challenging. There are many conducted studies to determine LA hemodynamic biomarkers using ECG signals, however, the low specificity has been seen as a common problem.

Purpose

Within the scope of this study, we aimed to develop a fully automatic artificial intelligence algorithm. Therefore, the frequency domain properties of the ECG signal were examined by wavelet transform and deep learning algorithm to detect the left atrial overload (LAO) from the ECG signal. We also compared the effectiveness of lead selection (Lead II and V1 Lead) to detect LAO.

Methods

Physionet-PTB-XL database was used for ECG inputs. 10 s ECG waveform of 403 healthy people and 352 LAO patients used for feature extraction after carefully filtering. Each signal was segmented to a 1.5 s epoch and in total 4530 ECG was used for training (70%) and test (30%) of deep learning algorithms. Two different deep learning models were developed for the automatic detection of LAO. In the first approach (1D convolutional neural network, 1D-CNN) segmented raw ECG was used, and in the second approach (2D convolutional neural network, 2D-CNN) each segmented ECG was transformed to scalograms with continuous wavelets transform. All models were tested for both leads.

Results

Our results show that frequency-domain features are highly capable to detect LAO compared to time-domain features and either Lead II and V1 Lead can be used for the detection of LAO. The success rate of 1D-CNN is 72% for Lead-II and 75% for V1 Lead while the success rate of 2D-CNN is 92% for both leads.

Conclusions

The most important limitation in the diagnosis of LAO by ECG is the low specificity. In this study, a model that eliminates this limitation has been developed by using the frequency characteristics of the ECG and artificial intelligence algorithms and tested with a large population. Considering its high success rate, 2D-CNN model has the potential to help the clinician by detecting the pathology with a cheaper, operator-independent method, and a short-term measurement.

Funding Acknowledgement

Type of funding sources: None.

The role of calcification in cardiovascular outcome after left main bifurcation revascularization: a single centre experience

Introduction

Left main bifurcation (LMB) disease represents a high-risk subset of advanced coronary artery disease, often associated with severe calcification. Different stenting techniques have been evaluated to overcome challenges of the LMB anatomy, but the role of the calcific burden on cardiovascular (CV) outcome after LMB revascularization is unclear.

Purpose

We sought to evaluate the CV hospitalization predictors during follow-up of patients who underwent LMB revascularization (LMBR) in a high-volume center in Italy.

Methods

We performed a retrospective analysis of LMBR patients between 2018 and 2021. Patients were treated with different techniques in the acute or chronic settings and followed-up by telephone or outpatient visit. Coronary calcification (CC) was visually judged as absent, mild, moderate or severe. Predictors of CV hospitalizations were assessed.

Results

The median follow-up was 511 days. Among 129 patients who underwent LMBR during the study period, 32 (24.8%) were female, with a mean age of 72 (±10.6) years. 105 (81.4%) patients were hypertensive and 49 (38%) diabetics. The mean eGFR value was 66.3 (± 21.8) ml/min/m2. The majority of patients had three vessel disease (79, 61.3%), while only a small minority had one vessel disease (5, 3.9%). The mean SYNTAX score I was 27.1 (± 8.6), with most patients at intermediate risk (65, 50.4%) followed by patients at low (37, 28.7%) and high (27, 20.9%) risk. A provisional technique was used in most cases (79, 61.2%), followed by double-kissing crush (37, 28.7%) and T-stent/T-and-protrusion (13, 10.1%). Intravascular imaging was used in 84 (65.1%) cases. Most patients had no angiographic demonstration of CC (73, 57.9%), while when present, they were mild in 14 (11.1%), moderate in 29 (23%) and severe in 10 (7.9%) patients. Preparation of the CC was performed only with non-compliant (NC) balloons in mild CC (10, 71.4%) and with NC balloons (14, 48.3%) or intravascular lithotripsy (IVL) (13, 44.8%) in case of moderate CC. In the presence of severe CC, lesion preparation was carried out with IVL (4, 40%) or NC balloons (6, 60%). CC were associated with a more unfavorable outcome and, when present in a severe grade, resulted in a statistically significant risk of CV hospitalizations (HR 1.652; 95% CI 1.723–15.793; p=0.003) (Figure 1). After univariate and multivariate Cox regression analysis (Figure 2), only the presence of severe CC was associated with an increased risk of CV hospitalizations (HR 1.9; 95% CI 1.76–19.63; p=0.002), whereas aspirin therapy was a protective factor (HR −1.34; CI 0.07–0.86; p=0.02).

Conclusions

The presence of severe calcification is associated with a higher risk of CV hospitalizations, despite preparation of calcific lesions was always performed and intravascular imaging use was extensive. There were no differences in outcomes regardless to clinical presentation at admission, different stenting techniques and SYNTAX score I.

Funding Acknowledgement

Type of funding sources: None.

Erythropoietin stimulates the coronary collateral development in patients with coronary chronic total occlusion

Objective

Erythropoietin (EPO) improves cardiac function and induces neovascularisation in post-myocardial infarction heart failure. The aim of this study was to analyse the association between the serum erythropoietin level and coronary collateral development in patients with coronary artery disease and chronic total occlusion.

Methods

A total of 168 patients consisting of 117 with coronary artery disease (CAD, (62 with chronic total occlusion (CTO), 55 without CTO)) and 51 with healthy coronary arteries were included in the study. The patients were assigned as coronary artery disease without CTO (group 0), CAD with CTO (group 1: poor collateral development, group 2: good collateral development) and normal coronary arteries (group 3).

Results

There was a significant positive correlation between serum EPO levels and the Rentrop scores in angiography (r = 0.243, p = 0.001). Similarly, a positive correlation was found between serum EPO levels and the Syntax scores (r = 0.253, p = 0.001). Echocardiography revealed a negative correlation between serum EPO levels and the cardiac ejection fraction (r = −0.210, p = 0.006).

Conclusions

Serum EPO is a useful biomarker for coronary collateral development in patients with CTO.