Transferable approach for cardiac disease classification using deep learning

Gopika, P.; Sowmya, V.; Gopalakrishnan, E. A.; Agarwal, Basant; Bălaş, Valentina Emilia; Jain, Lakhmi C.; Poonia, Ramesh Chandra; Manisha, .

doi:10.1016/b978-0-12-819061-6.00012-4

Cited by 19 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Existing studies in classification of cardiac diseases are disease specific based approaches. On the contrary, [22] determined single best architecture in classifying heart ailments like arrhythmia, atrial fibrillation and heart attack. Architectures such as CNN, RNN, LSTM and GRU compared and arrived at a transferable approach for hyperparameters in disease classification.…”

Section: Deep Learning-based Approaches For Heart Diseases Detectionmentioning

confidence: 97%

“…Testing methods on a large set of data obtained clinically and lack of sophisticated method in removing irrelevant features was not clearly presented in the literature [6] because of its missing values and noise management interrupted efficient results [2]. Transferable approach proposed by [22] did not take into account pathological measures such as phonocardiogram in diagnosing cardiac diseases. Li et al [38] proposed method had drawbacks as it was not a comprehensive framework in early diagnosis of cardiovascular diseases.…”

Section: Research Gapsmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Literature Review on Automatic Recognition and Classification of Coronary Atherosclerosis

Prajapati¹,

Patel²

2022

Artificial Intelligence and Communication Technologies

View full text Add to dashboard Cite

Cardiovascular diseases have a high morbidity rate and per year it leads to 17 million deaths worldwide. Coronary Atherosclerosis is one of the chief causes of stroke and progressive heart disease marked by lipids and fibrous elements accumulation in the arteries. Artificial intelligence (AI) has established remarkable progress in recent times in clinical practice helping patients and healthcare professionals in the accurate and faster diagnosis of diseases. Prediction models in the identification of Atherosclerosis have set foot in the academic literature to assist in making medical decisions during urgent circumstances. This paper aimed to systematically review automatic atherosclerotic plaque detection algorithms. The advantages of these latest techniques in automatic recognition of atherosclerotic plaque, its composition, classification strategy and future predictions in terms of severity are elucidated in this review with limitations and research gaps. The findings suggest that deep learning models are the future of diagnosis and ensemble learning algorithms are best in non-invasive accurate detection of cardiovascular diseases.

show abstract

Section: Deep Learning-based Approaches For Heart Diseases Detectionmentioning

confidence: 97%

Section: Research Gapsmentioning

confidence: 99%

A Systematic Literature Review on Automatic Recognition and Classification of Coronary Atherosclerosis

Prajapati¹,

Patel²

2022

Artificial Intelligence and Communication Technologies

View full text Add to dashboard Cite

show abstract

“…In terms of cardiovascular safety and patient outcomes, it's also important to create a 'evidencebaseline' for large-data applications. Furthermore, advances in sensor technology made it possible to reliably translate more aspects of reality into data in small packages while using less energy and spending less money [14]. When taken as a whole, using data today is much less costly than it has ever been.…”

Section: Big Data In Cardiologymentioning

confidence: 99%

Machine Learning Era In Heart Disease Prediction- An Intense Learning Analysis With Big Data

Prasad¹

2021

TURCOMAT

View full text Add to dashboard Cite

Machine learning and adaptation is a collection of machine learning methods consisting of several stacked layers and using data to explore hierarchical abstractions. As computer power has increased and large data has emerged, deep learning is an appropriate structure for cardiological tasks. The need to optimize medical treatment varies from diagnostic to therapeutic in the absence of a medical Centre. Machine learning systems are previous attempts to imitate medical practitioners in their protocol for solving medical tasks or for producing observations. These systems are known not to be useful as they require extensive design features and domain expertise in order to achieve the new cardio data highly accurate and difficult to map. Overall, with any technical progress, cardiometry and medicine are autonomous and become closer to an automated, detailed learning area. But no complete conceptual basis for in-depth education can be found. A thorough analysis of its internal functional qualities and constraints is required to enable the field to adopt its position on the disease of the heart.In this study, a large number of very complex machine learning concepts integrated into the cardio domain with big data have been studied over a very short time period of time.

show abstract

“…In the previous study, the ECG signal classification based on heart rhythm can be conducted with several features morphology of ECG signal like presenting ST-elevation and depression, T-wave abnormalities, and pathological Q-waves ( Ansari et al, 2017 ). Moreover, a variety of ECG rhythm features, such as the R-R interval, S-T interval, P-R interval, and Q-T interval have been implemented to automatically detect heart abnormalities over the past decade ( Gopika et al, 2020 ). Unlike an ECG rhythm, the efficiency classification of the irregular heartbeat , either faster or slower than normal, or even waveform malformation can be improve by using beat feature ( Khalaf, Owis & Yassine, 2015 ).…”

Section: Introductionmentioning

confidence: 99%

Deep learning-based electrocardiogram rhythm and beat features for heart abnormality classification

Darmawahyuni

Nurmaini

Rachmatullah

et al. 2022

PeerJ Computer Science

View full text Add to dashboard Cite

Background Electrocardiogram (ECG) signal classification plays a critical role in the automatic diagnosis of heart abnormalities. While most ECG signal patterns cannot be recognized by a human interpreter, they can be detected with precision using artificial intelligence approaches, making the ECG a powerful non-invasive biomarker. However, performing rapid and accurate ECG signal classification is difficult due to the low amplitude, complexity, and non-linearity. The widely-available deep learning (DL) method we propose has presented an opportunity to substantially improve the accuracy of automated ECG classification analysis using rhythm or beat features. Unfortunately, a comprehensive and general evaluation of the specific DL architecture for ECG analysis across a wide variety of rhythm and beat features has not been previously reported. Some previous studies have been concerned with detecting ECG class abnormalities only through rhythm or beat features separately. Methods This study proposes a single architecture based on the DL method with one-dimensional convolutional neural network (1D-CNN) architecture, to automatically classify 24 patterns of ECG signals through both rhythm and beat. To validate the proposed model, five databases which consisted of nine-class of ECG-base rhythm and 15-class of ECG-based beat were used in this study. The proposed DL network was applied and studied with varying datasets with different frequency samplings in intra and inter-patient scheme. Results Using a 10-fold cross-validation scheme, the performance results had an accuracy of 99.98%, a sensitivity of 99.90%, a specificity of 99.89%, a precision of 99.90%, and an F1-score of 99.99% for ECG rhythm classification. Additionally, for ECG beat classification, the model obtained an accuracy of 99.87%, a sensitivity of 96.97%, a specificity of 99.89%, a precision of 92.23%, and an F1-score of 94.39%. In conclusion, this study provides clinicians with an advanced methodology for detecting and discriminating heart abnormalities between different ECG rhythm and beat assessments by using one outstanding proposed DL architecture.

show abstract

Transferable approach for cardiac disease classification using deep learning

Cited by 19 publications

References 2 publications

A Systematic Literature Review on Automatic Recognition and Classification of Coronary Atherosclerosis

A Systematic Literature Review on Automatic Recognition and Classification of Coronary Atherosclerosis

Machine Learning Era In Heart Disease Prediction- An Intense Learning Analysis With Big Data

Deep learning-based electrocardiogram rhythm and beat features for heart abnormality classification

Contact Info

Product

Resources

About