Assessment of qualitative and quantitative features in coronary artery MRA

Heart failure classification using deep learning to extract spatiotemporal features from ECG

BACKGROUND

Heart failure is a syndrome with complex clinical manifestations. Due to increasing population aging, heart failure has become a major medical problem worldwide. In this study, we used the MIMIC-III public database to extract the temporal and spatial characteristics of electrocardiogram (ECG) signals from patients with heart failure.

METHODS

We developed a NYHA functional classification model for heart failure based on a deep learning method. We introduced an integrating attention mechanism based on the CNN-LSTM-SE model, segmenting the ECG signal into 2 to 20 s long segments. Ablation experiments showed that the 12 s ECG signal segments could be used with the proposed deep learning model for superior classification of heart failure.

RESULTS

The accuracy, positive predictive value, sensitivity, and specificity of the NYHA functional classification method were 99.09, 98.9855, 99.033, and 99.649%, respectively.

CONCLUSIONS

The comprehensive performance of this model exceeds similar methods and can be used to assist in clinical medical diagnoses.

A new diagnostic autism spectrum disorder (DASD) strategy using ensemble diagnosis methodology based on blood tests

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disease that impacts a child's way of behavior and social communication. In early childhood, children with ASD typically exhibit symptoms such as difficulty in social interaction, limited interests, and repetitive behavior. Although there are symptoms of ASD disease, most people do not understand these symptoms and therefore do not have enough knowledge to determine whether or not a child has ASD. Thus, early detection of ASD children based on accurate diagnosis model based on Artificial Intelligence (AI) techniques is a critical process to reduce the spread of the disease and control it early. Through this paper, a new Diagnostic Autism Spectrum Disorder (DASD) strategy is presented to quickly and accurately detect ASD children. DASD contains two layers called Data Filter Layer (DFL) and Diagnostic Layer (DL). Feature selection and outlier rejection processes are performed in DFL to filter the ASD dataset from less important features and incorrect data before using the diagnostic or detection method in DL to accurately diagnose the patients. In DFL, Binary Gray Wolf Optimization (BGWO) technique is used to select the most significant set of features while Binary Genetic Algorithm (BGA) technique is used to eliminate invalid training data. Then, Ensemble Diagnosis Methodology (EDM) as a new diagnostic technique is used in DL to quickly and precisely diagnose ASD children. In this paper, the main contribution is EDM that consists of several diagnostic models including Enhanced K-Nearest Neighbors (EKNN) as one of them. EKNN represents a hybrid technique consisting of three methods called K-Nearest Neighbors (KNN), Naïve Bayes (NB), and Chimp Optimization Algorithm (COA). NB is used as a weighed method to convert data from feature space to weight space. Then, COA is used as a data generation method to reduce the size of training dataset. Finally, KNN is applied on the reduced data in weight space to quickly and accurately diagnose ASD children based on new training dataset with small size. ASD blood tests dataset is used to test the proposed DASD strategy against other recent strategies [1]. It is concluded that the DASD strategy is superior to other strategies based on many performance measures including accuracy, error, recall, precision, micro_average precision, macro_average precision, micro_average recall, macro_average recall, F1-measure, and implementation-time with values equal to 0.93, 0.07, 0.83, 0.82, 0.80, 0.83, 0.79, 0.81, 0.79, and 1.5 s respectively.

Arrhythmia classification detection based on multiple electrocardiograms databases

According to the World Health Organization, cardiovascular diseases are the leading cause of deaths globally. Electrocardiogram (ECG) is a non-invasive approach for detecting heart diseases and reducing the risk of heart disease-related death. However, there are limited numbers of ECG samples and imbalance distribution for existing ECG databases. It is difficult to train practical and efficient neural networks. Based on the analysis and research of many existing ECG databases, this paper conduct an in-depth study on three fine-labeled ECG databases, to extract heartbeats, unify the sampling frequency, and propose a self-processing method of heartbeats, and finally form a unified ECG arrhythmia classification database, noted as Hercules-3. It is separated into training sets (80%) and testing sets (the remaining 20%). In order to verify its capabilities, we have trained a 16-classification fully connected neural network based on Hercules-3 and it achieves an accuracy rate of up to 98.67%. Compared with other data processing, our proposed method improves classification recall by at least 6%, classification accuracy by at least 4%, and F1-score by at least 7%.

Accuracy of intravascular ultrasound-derived virtual fractional flow reserve (FFR) and FFR derived from computed tomography for functional assessment of coronary artery disease

BACKGROUND

Coronary computed tomography-derived fractional flow reserve (CT-FFR) and intravascular ultrasound-derived fractional flow reserve (IVUS-FFR) are two functional assessment methods for coronary stenoses. However, the calculation algorithms for these methods differ significantly. This study aimed to compare the diagnostic performance of CT-FFR and IVUS-FFR using invasive fractional flow reserve (FFR) as the reference standard.

METHODS

Six hundred and seventy patients (698 lesions) with known or suspected coronary artery disease were screened for this retrospective analysis between January 2020 and July 2021. A total of 40 patients (41 lesions) underwent intravascular ultrasound (IVUS) and FFR evaluations within six months after completing coronary CT angiography were included. Two novel CFD-based models (AccuFFRct and AccuFFRivus) were used to compute the CT-FFR and IVUS-FFR values, respectively. The invasive FFR ≤ 0.80 was used as the reference standard for evaluating the diagnostic performance of CT-FFR and IVUS-FFR.

RESULTS

Both AccuFFRivus and AccuFFRct demonstrated a strong correlation with invasive FFR (R = 0.7913, P < 0.0001; and R = 0.6296, P < 0.0001), and both methods showed good agreement with FFR. The area under the receiver operating characteristic curve was 0.960 (P < 0.001) for AccuFFRivus and 0.897 (P < 0.001) for AccuFFRct in predicting FFR ≤ 0.80. FFR ≤ 0.80 were predicted with high sensitivity (96.6%), specificity (85.7%), and the Youden index (0.823) using the same cutoff value of 0.80 for AccuFFRivus. A good diagnostic performance (sensitivity 89.7%, specificity 85.7%, and Youden index 0.754) was also demonstrated by AccuFFRct.

CONCLUSIONS

AccuFFRivus, computed from IVUS images, exhibited a high diagnostic performance for detecting myocardial ischemia. It demonstrated better diagnostic power than AccuFFRct, and could serve as an accurate computational tool for ischemia diagnosis and assist in clinical decision-making.

Machine learning-enhanced echocardiography for screening coronary artery disease

BACKGROUND

Since myocardial work (MW) and left atrial strain are valuable for screening coronary artery disease (CAD), this study aimed to develop a novel CAD screening approach based on machine learning-enhanced echocardiography.

METHODS

This prospective study used data from patients undergoing coronary angiography, in which the novel echocardiography features were extracted by a machine learning algorithm. A total of 818 patients were enrolled and randomly divided into training (80%) and testing (20%) groups. An additional 115 patients were also enrolled in the validation group.

RESULTS

The superior diagnosis model of CAD was optimized using 59 echocardiographic features in a gradient-boosting classifier. This model showed that the value of the receiver operating characteristic area under the curve (AUC) was 0.852 in the test group and 0.834 in the validation group, with high sensitivity (0.952) and low specificity (0.691), suggesting that this model is very sensitive for detecting CAD, but its low specificity may increase the high false-positive rate. We also determined that the false-positive cases were more susceptible to suffering cardiac events than the true-negative cases.

CONCLUSIONS

Machine learning-enhanced echocardiography can improve CAD detection based on the MW and left atrial strain features. Our developed model is valuable for estimating the pre-test probability of CAD and screening CAD patients in clinical practice.

TRIAL REGISTRATION

Registered as NCT03905200 at ClinicalTrials.gov. Registered on 5 April 2019.

Artificial intelligence based approach for categorization of COVID-19 ECG images in presence of other cardiovascular disorders

Coronavirus disease (COVID-19) is a class of SARS-CoV-2 virus which is initially identified in the later half of the year 2019 and then evolved as a pandemic. If it is not identified in the early stage then the infection and mortality rates increase with time. A timely and reliable approach for COVID-19 identification has become important in order to prevent the disease from spreading rapidly. In recent times, many methods have been suggested for the detection of COVID-19 disease have various flaws, to increase diagnosis performance, fresh investigations are required. In this article, automatically diagnosing COVID-19 using ECG images and deep learning approaches like as Visual Geometry Group (VGG) and AlexNet architectures have been proposed. The proposed method is able to classify between COVID-19, myocardial infarction, normal sinus rhythm, and other abnormal heart beats using Lead-II ECG image only. The efficacy of the technique proposed is validated by using a publicly available ECG image database. We have achieved an accuracy of 77.42% using Alexnet model and 75% accuracy with the help of VGG19 model.

Critical evaluation of stents in coronary angioplasty: a systematic review

BACKGROUND

Coronary stents are routinely placed in the treatment and prophylaxis of coronary artery disease (CAD). Current coronary stent designs are prone to developing blockages: in-stent thrombosis (IST) and in-stent re-stenosis (ISR). This is a systematic review of the design of current coronary stent models, their structural properties and their modes of application, with a focus on their associated risks of IST and ISR. The primary aim of this review is to identify the best stent design features for reducing the risk of IST and ISR. To review the three major types of stents used in clinical settings today, determining best and relevant clinical practice by exploring which types and features of offer improved patient outcomes regarding coronary angioplasty. This information can potentially be used to increase the success rate of coronary angioplasty and stent technology in the future taking into account costs and benefits.

METHODS

Scientific databases were searched to find studies concerning stents. After the exclusion criteria were applied, 19 of the 3192 searched literature were included in this review. Studies investigating three major types of stent design were found: bare-metal stents (BMS), drug-eluting stents (DES) and bioresorbable stents (BRS). The number of participants varied between 14 and 1264. On average 77.4% were male, with a mean age of 64 years.

RESULTS

From the findings of these studies, it is clear that DES are superior in reducing the risk of ISR when compared to BMS. Conflicting results do not clarify whether BRS are superior to DES at reducing IST occurrence, although studies into newer BRS technologies show reducing events of IST to 0, creating a promising future for BRS showing them to be non-inferior. Thinner stents were shown to reduce IST rates, due to better re-endothelialisation. Scaffold material has also been shown to play a role with cobalt alloy stents reducing the risk of IST. This study found that thinner stents that release drugs were better at preventing re-blockages. Some dissolvable stents might be better at stopping blood clots blocking the arteries when compared to metal stents. The method and procedure of implanting the stent during coronary angioplasty influences success rate of these stents, meaning stent design is not the only significant factor to consider.

CONCLUSIONS

Positive developments in coronary angioplasty could be made by designing new stents that encompass all the most desirable properties of existing stent technology. Further work is needed to investigate the benefits of BRS in reducing the risk of IST compared to DES, as well as to investigate the effects of different scaffold materials on IST and ISR outcomes.

Understanding the mechanisms amenable to CRT response: from pre-operative multimodal image data to patient-specific computational models

This manuscript describes our recent developments towards better understanding of the mechanisms amenable to cardiac resynchronization therapy response. We report the results from a full multimodal dataset corresponding to eight patients from the euHeart project. The datasets include echocardiography, MRI and electrophysiological studies. We investigate two aspects. The first one focuses on pre-operative multimodal image data. From 2D echocardiography and 3D tagged MRI images, we compute atlas based dyssynchrony indices. We complement these indices with presence and extent of scar tissue and correlate them with CRT response. The second one focuses on computational models. We use pre-operative imaging to generate a patient-specific computational model. We show results of a fully automatic personalized electromechanical simulation. By case-per-case discussion of the results, we highlight the potential and key issues of this multimodal pipeline for the understanding of the mechanisms of CRT response and a better patient selection.