Time-Domain Analysis of the Electrocardiogram

Chouvarda, Ioanna; Filos, Dimitrios; Maglaveras, Nicos

doi:10.1007/978-981-10-5092-3_4

Cited by 3 publications

(2 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first baseline that we evaluated was the model developed using ECG time-domain features which is a popular way of extracting information from the ECG waveforms with low computational complexity (Chouvarda et al 2019). The ECG signal is normalized between [−1,1] before feature extraction and to remove baseline wander, we calculated the median value of the ECG signal window and subtracted it from all the samples in that window.…”

Section: Baseline 1 a Predictive Model Using Time-domain Analysis Of ...mentioning

confidence: 99%

Multi-modal fusion model for predicting adverse cardiovascular outcome post percutaneous coronary intervention

Bhattacharya¹,

Sadasivuni²,

Chao³

et al. 2022

Physiol. Meas.

View full text Add to dashboard Cite

Background. Clinical medicine relies heavily on the synthesis of information and data from multiple sources. However, often simple feature concatenation is used as a strategy for developing a multimodal machine learning model in the cardiovascular domain, and thus the models are often limited by pre-selected features and moderate accuracy. Method. We proposed a two-branched joint fusion model for fusing the 12-lead electrocardiogram (ECG) signal data with clinical variables from the electronic medical record (EMR) in an end-to-end deep learning architecture. The model follows the joint fusion scheme and learns complementary information from ECG and EMR. Retrospective data from the Mayo Clinic Health Systems across four sites (La Crosse, WI; Mankato, MN; Rochester, MN; Scottsdale, AZ) for patients that underwent percutaneous coronary intervention (PCI) between January 2006 and December 2018 were obtained. Model performance was assessed by area under the receiver-operating characteristics (AUROC) and Delong’s test. Results. The final cohort included 17,356 unique patients with a mean age of 67.2 ± 12.6 and 9,163 (52.7%) were male. The joint fusion model outperformed the ECG time-domain model with statistical margin. The model with clinical data obtained the highest AUROC for all-cause mortality (0.91 at 6 months) but the joint fusion model outperformed for cardiovascular outcomes - heart failure hospitalization and ischemic stroke with a significant margin (Delong’s p < 0.05). Conclusion. To the best of our knowledge, this is the first study that developed a deep learning model with joint fusion architecture for the prediction of post-PCI prognosis and outperformed machine learning models developed using traditional single-source features (clinical variables or ECG features). Adding ECG data with clinical variables did not improve prediction of all-cause mortality as may be expected, but the improved performance of related cardiac outcomes shows that the fusion of ECG generates additional value.

show abstract

Section: Baseline 1 a Predictive Model Using Time-domain Analysis Of ...mentioning

confidence: 99%

Multi-modal fusion model for predicting adverse cardiovascular outcome post percutaneous coronary intervention

Bhattacharya¹,

Sadasivuni²,

Chao³

et al. 2022

Physiol. Meas.

View full text Add to dashboard Cite

show abstract

“…When a band-pass filter is used, however, noise removal is not efficient and detection accuracy is low. To detect with high accuracy, ECG morphology [6,7], time-frequency domains [8][9][10], and genetic algorithms [11,12] are mainly used. ECG morphology technique utilizes a neural network based approach.…”

Section: Related Workmentioning

confidence: 99%

Development of a Knowledge Discovery Computing based wearable ECG monitoring system

2020

View full text Add to dashboard Cite

ECG signals contain a lot of information related to cardiac activity and play a critical role in diagnosing of heart disease. However, relying on health information monitoring with simple ECG measurement can lead to errors in health information analysis. Therefore, for accurate diagnosis and analysis, an algorithm that can efficiently process user's measurement environment and activity information is required. In this paper, we propose a wearable ECG monitoring system based on Knowledge Discovery Computing using 3-axis acceleration sensor. The proposed system measures real-time cardiac information and activity information simultaneously to minimize errors in health information analysis through Knowledge Discovery Computing between the user's environment information and abnormal ECG according to the measurement environment in everyday life. In addition, we implemented a packet transmission protocol to effectively transmit large amounts of data analyzed through Knowledge Discovery Computing to the base station. First, arrhythmia detection was performed using R-peak detection preprocessing and pattern matching algorithm. Also, a classification algorithm was implemented to classify activity types by utilizing an accelerometer in order to recognize the context surrounding the user. Information on the user's vital signs and activity information can be used for more accurately determine arrhythmia in daily life. Also, variable packet generation protocol was designed for an effective transmission of data packets increased exponentially by long-term measurements and wireless data transfer. The variable packet generation protocol is efficient in limited wireless network environments, because it generate packets of the entire data only with case of abnormal cardiac rhythm and transmits minimal information for normal cardiac rhythm. In order to evaluate the performance of ECG monitoring system based on Knowledge Discovery Computing, we designed a 2-lead ECG measurement belt manufactured with conductive fiber to minimize user discomfort, and assessed the system performance in data packet transmission, data recovery, and arrhythmia detection in dynamic states in daily life. In static states, the posture detection was 100%, heart rate detection 99.8%, and CR (Compression Ratio) was 193.99 and correlation coefficient of 0.95 with commercial systems. In dynamic states, 96% detection rate and 59.14 of CR is identified. If arrhythmia is determined based only on ECG signals, it is difficult to differentiate an actual abnormal cardiac rhythm from an ECG signal altered due to motion. The experiments conducted in this study confirmed that Knowledge Discovery is possible in the dynamic state through the proposed system in daily life.

show abstract

Transformer-Based Network for Accurate Classification of Lung Auscultation Sounds

Sonali,

Kiran,

Chinmayi

et al. 2023

Crit Rev Biomed Eng

View full text Add to dashboard Cite

Respiratory diseases are a major cause of death worldwide, affecting a significant proportion of the population with lung function abnormalities that can lead to respiratory illnesses. Early detection and prevention are critical to effective management of these disorders. Deep learning algorithms offer a promising approach for analyzing complex medical data and aiding in early disease detection. While transformer-based models for sequence classification have proven effective for tasks like sentiment analysis, topic classification, etc., their potential for respiratory disease classification remains largely unexplored. This paper proposes a classifier utilizing the transformer-encoder block, which can capture complex patterns and dependencies in medical data. The proposed model is trained and evaluated on a large dataset from the International Conference on Biomedical Health Informatics 2017, achieving state-of-the-art results with a mean sensitivity of 70.53%, mean specificity of 84.10%, mean average score of 77.32%, and mean harmonic score of 76.10%. These results demonstrate the model's effectiveness in diagnosing respiratory diseases while taking up minimal computational resources.

show abstract

Time-Domain Analysis of the Electrocardiogram

Cited by 3 publications

References 79 publications

Multi-modal fusion model for predicting adverse cardiovascular outcome post percutaneous coronary intervention

Multi-modal fusion model for predicting adverse cardiovascular outcome post percutaneous coronary intervention

Development of a Knowledge Discovery Computing based wearable ECG monitoring system

Transformer-Based Network for Accurate Classification of Lung Auscultation Sounds

Contact Info

Product

Resources

About