Machine Learning Methods in Predicting Patients with Suspected Myocardial Infarction Based on Short-Time HRV Data

Chumachenko, Dmytro; Butkevych, Mykola; Lode, Daniel; Frohme, Marcus; Schmailzl, Kurt J. G.; Nechyporenko, Alina

doi:10.3390/s22187033

Cited by 15 publications

(6 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our research delved into classifying five distinct heart rhythm types from the MIT-BIH database, deploying diverse machine learning ML and deep learning DL algorithms. Examples of ML algorithms used here are as follows: principal component analysis (PCA), which scrutinizes the ECG signal's complexities, identifying key patterns and reducing its dimensionality [39,40]; independent component analysis (ICA) which unmasks hidden or overlapping signals lurking within the ECG [41,42]; random forest (RF) which builds a team of decision trees, each analyzing the ECG signal slightly differently, and then votes for the most likely heart rhythm category [43][44][45]; and K-best algorithm that picks the K most relevant features from the ECG signal, focusing on the distinctive features that matter most for classification [46]. As for deep learning algorithms, both CNN and LSTM were used.…”

Section: Results Of the First Scenariomentioning

confidence: 99%

Electrocardiogram Signals Classification Using Deep-Learning-Based Incorporated Convolutional Neural Network and Long Short-Term Memory Framework

Eleyan,

Alboghbaish

2024

Computers

View full text Add to dashboard Cite

Cardiovascular diseases (CVDs) like arrhythmia and heart failure remain the world’s leading cause of death. These conditions can be triggered by high blood pressure, diabetes, and simply the passage of time. The early detection of these heart issues, despite substantial advancements in artificial intelligence (AI) and technology, is still a significant challenge. This research addresses this hurdle by developing a deep-learning-based system that is capable of predicting arrhythmias and heart failure from abnormalities in electrocardiogram (ECG) signals. The system leverages a model that combines long short-term memory (LSTM) networks with convolutional neural networks (CNNs). Extensive experiments were conducted using ECG data from both the MIT-BIH and BIDMC databases under two scenarios. The first scenario employed data from five distinct ECG classes, while the second focused on classifying data from three classes. The results from both scenarios demonstrated that the proposed deep-learning-based classification approach outperformed existing methods.

show abstract

Section: Results Of the First Scenariomentioning

confidence: 99%

Electrocardiogram Signals Classification Using Deep-Learning-Based Incorporated Convolutional Neural Network and Long Short-Term Memory Framework

Eleyan,

Alboghbaish

2024

Computers

View full text Add to dashboard Cite

show abstract

“…HRV was also monitored. The time and frequency domains were generated automatically using Medeia 3000 including high frequency (HF), low frequency (LF), SDNN, and PNN50 [11]. Carotid plaque detection was tested by the Korea GE Color Ultrasound Diagnostic Instrument (LOGIQ S7).…”

Section: Clinical and Biochemical Measurementsmentioning

confidence: 99%

Analysis of gastric electrical rhythm in patients with type 2 diabetes mellitus

Wang,

Ma,

Jiang

et al. 2024

Endocrine

View full text Add to dashboard Cite

Aim To analysis the change of electrogastrogram (EGG) in patients with type 2 diabetes mellitus (T2DM), and evaluate the prevalence of abnormal gastric electrical rhythm (AGER) and its relative influencing factors. Methods A total of 65 patients with T2DM hospitalized at the Second Affiliated Hospital of Soochow University from Dec. 2020 to Dec. 2021 were included in the cross-sectional study. General information, clinical data, and medical history data of all study subjects, including name, gender, body mass index (BMI), duration of diabetes, anti-diabetic therapies, high blood pressure (HBP) history, smoking history, and medication history, were completely collected. The results of laboratory tests, including biochemical parameters, glycosylated hemoglobin (HbA1c), fasting C-peptide, 2 h postprandial C-peptide, 24 h urine total protein (24 hUTP), urine microalbumin creatinine ratio (UACR) and estimated glomerular filtration rate (eGFR) were recorded. EGG, Gastroparesis Cardinal Symptom Index (GCSI), gastric emptying ultrasound, fundus examination, carotid artery ultrasonography, cardiac autonomic function test, heart rate variability (HRV) were all examined and recorded as well. According to the results of EGG, the subjects were divided into normal gastric electrical rhythm (NGER) group and abnormal gastric electrical rhythm (AGER) group. Results (1) Fasting blood glucose (FBG), HbA1c, the presence of diabetic peripheral neuropathy (DPN) and diabetic cardiac autonomic neuropathy (DCAN) were significantly higher in the AGER group (p < 0.05). Low frequency (LF) and high frequency (HF), the indicators of HRV, were significantly lower in the AGER group (p < 0.05). In addition, the prevalence of feeling excessively full after meals, loss of appetite, and stomach or belly visibly larger after meals of gastrointestinal symptoms of gastroparesis were significantly higher in the AGER group (p < 0.05). Multiple logistic regression analysis showed that FBG and the prevalence of DCAN were the independent risk factors. Conclusion AGER was associated with high FBG and the presence of DCAN. EGG examination is recommended for patients with gastrointestinal symptoms and clues of DCAN.

show abstract

“…In addition, the algorithm is trustworthy. Typically, a new data point may only have a minor effect on one tree, therefore the algorithm as a whole is not significantly affected [23]. This algorithm is fulfilled through the below steps:…”

Section: Random Forestmentioning

confidence: 99%

Secure Heart Disease Classification System Based on Three Pass Protocol and Machine Learning

shaker

Obayes

Al-Shareefi

2023

ijcsm

View full text Add to dashboard Cite

Heart disease is one of the worst life-threatening conditions. Correct and early diagnosis of this disease is crucial for saving patients’ life and avoiding other complications. On the other hand, keeping the patient’s data, diagnosis process, and treatment plan secured is equally important to the defactomedical procedure. This research proposes a system that is consisting of two phases: security provision and patients’ condition diagnosis. Typically, the first phase exercises a security protocol, called three-pass protocol, to ensure that the people who can access the patient's information are authorized. In order to obtain a high accuracy level in the diagnosis process, artificial intelligence with machine learning methods are employed in the later phase. The proposed system relies on a data set which includes a number of vital indicators, by which the patient's status can be classified as having heart disease or not. The KNN algorithm and the random forest tree algorithm are applied to carry out the classification task. The accuracy scale results reveals that the randomforest tree algorithm (99%) gave higher accuracy than KNN (97%).

show abstract

Machine Learning Methods in Predicting Patients with Suspected Myocardial Infarction Based on Short-Time HRV Data

Cited by 15 publications

References 47 publications

Electrocardiogram Signals Classification Using Deep-Learning-Based Incorporated Convolutional Neural Network and Long Short-Term Memory Framework

Electrocardiogram Signals Classification Using Deep-Learning-Based Incorporated Convolutional Neural Network and Long Short-Term Memory Framework

Analysis of gastric electrical rhythm in patients with type 2 diabetes mellitus

Secure Heart Disease Classification System Based on Three Pass Protocol and Machine Learning

Contact Info

Product

Resources

About