scholarly journals Automated Detection of Acute Myocardial Infarction Using Asynchronous Electrocardiogram Signals—Preview of Implementing Artificial Intelligence With Multichannel Electrocardiographs Obtained From Smartwatches: Retrospective Study

10.2196/31129 ◽  
2021 ◽  
Vol 23 (9) ◽  
pp. e31129
Author(s):  
Changho Han ◽  
Youngjae Song ◽  
Hong-Seok Lim ◽  
Yunwon Tae ◽  
Jong-Hwan Jang ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
The Body ◽  
Diagnostic Code ◽  
Analysis Software ◽  
Ecg Analysis ◽  
Ecg Signals ◽  
Cardiac Disorders ◽  
Ecg Leads

Background When using a smartwatch to obtain electrocardiogram (ECG) signals from multiple leads, the device has to be placed on different parts of the body sequentially. The ECG signals measured from different leads are asynchronous. Artificial intelligence (AI) models for asynchronous ECG signals have barely been explored. Objective We aimed to develop an AI model for detecting acute myocardial infarction using asynchronous ECGs and compare its performance with that of the automatic ECG interpretations provided by a commercial ECG analysis software. We sought to evaluate the feasibility of implementing multiple lead–based AI-enabled ECG algorithms on smartwatches. Moreover, we aimed to determine the optimal number of leads for sufficient diagnostic power. Methods We extracted ECGs recorded within 24 hours from each visit to the emergency room of Ajou University Medical Center between June 1994 and January 2018 from patients aged 20 years or older. The ECGs were labeled on the basis of whether a diagnostic code corresponding to acute myocardial infarction was entered. We derived asynchronous ECG lead sets from standard 12-lead ECG reports and simulated a situation similar to the sequential recording of ECG leads via smartwatches. We constructed an AI model based on residual networks and self-attention mechanisms by randomly masking each lead channel during the training phase and then testing the model using various targeting lead sets with the remaining lead channels masked. Results The performance of lead sets with 3 or more leads compared favorably with that of the automatic ECG interpretations provided by a commercial ECG analysis software, with 8.1%-13.9% gain in sensitivity when the specificity was matched. Our results indicate that multiple lead-based AI-enabled ECG algorithms can be implemented on smartwatches. Model performance generally increased as the number of leads increased (12-lead sets: area under the receiver operating characteristic curve [AUROC] 0.880; 4-lead sets: AUROC 0.858, SD 0.008; 3-lead sets: AUROC 0.845, SD 0.011; 2-lead sets: AUROC 0.813, SD 0.018; single-lead sets: AUROC 0.768, SD 0.001). Considering the short amount of time needed to measure additional leads, measuring at least 3 leads—ideally more than 4 leads—is necessary for minimizing the risk of failing to detect acute myocardial infarction occurring in a certain spatial location or direction. Conclusions By developing an AI model for detecting acute myocardial infarction with asynchronous ECG lead sets, we demonstrated the feasibility of multiple lead-based AI-enabled ECG algorithms on smartwatches for automated diagnosis of cardiac disorders. We also demonstrated the necessity of measuring at least 3 leads for accurate detection. Our results can be used as reference for the development of other AI models using sequentially measured asynchronous ECG leads via smartwatches for detecting various cardiac disorders.

scholarly journals Automated Detection of Acute Myocardial Infarction Using Asynchronous Electrocardiogram Signals—Preview of Implementing Artificial Intelligence With Multichannel Electrocardiographs Obtained From Smartwatches: Retrospective Study (Preprint)

2021 ◽  
Author(s):  
Changho Han ◽  
Youngjae Song ◽  
Hong-Seok Lim ◽  
Yunwon Tae ◽  
Jong-Hwan Jang ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
The Body ◽  
Diagnostic Code ◽  
Analysis Software ◽  
Ecg Analysis ◽  
Ecg Signals ◽  
Cardiac Disorders ◽  
Ecg Leads

BACKGROUND When using a smartwatch to obtain electrocardiogram (ECG) signals from multiple leads, the device has to be placed on different parts of the body sequentially. The ECG signals measured from different leads are asynchronous. Artificial intelligence (AI) models for asynchronous ECG signals have barely been explored. OBJECTIVE We aimed to develop an AI model for detecting acute myocardial infarction using asynchronous ECGs and compare its performance with that of the automatic ECG interpretations provided by a commercial ECG analysis software. We sought to evaluate the feasibility of implementing multiple lead–based AI-enabled ECG algorithms on smartwatches. Moreover, we aimed to determine the optimal number of leads for sufficient diagnostic power. METHODS We extracted ECGs recorded within 24 hours from each visit to the emergency room of Ajou University Medical Center between June 1994 and January 2018 from patients aged 20 years or older. The ECGs were labeled on the basis of whether a diagnostic code corresponding to acute myocardial infarction was entered. We derived asynchronous ECG lead sets from standard 12-lead ECG reports and simulated a situation similar to the sequential recording of ECG leads via smartwatches. We constructed an AI model based on residual networks and self-attention mechanisms by randomly masking each lead channel during the training phase and then testing the model using various targeting lead sets with the remaining lead channels masked. RESULTS The performance of lead sets with 3 or more leads compared favorably with that of the automatic ECG interpretations provided by a commercial ECG analysis software, with 8.1%-13.9% gain in sensitivity when the specificity was matched. Our results indicate that multiple lead-based AI-enabled ECG algorithms can be implemented on smartwatches. Model performance generally increased as the number of leads increased (12-lead sets: area under the receiver operating characteristic curve [AUROC] 0.880; 4-lead sets: AUROC 0.858, SD 0.008; 3-lead sets: AUROC 0.845, SD 0.011; 2-lead sets: AUROC 0.813, SD 0.018; single-lead sets: AUROC 0.768, SD 0.001). Considering the short amount of time needed to measure additional leads, measuring at least 3 leads—ideally more than 4 leads—is necessary for minimizing the risk of failing to detect acute myocardial infarction occurring in a certain spatial location or direction. CONCLUSIONS By developing an AI model for detecting acute myocardial infarction with asynchronous ECG lead sets, we demonstrated the feasibility of multiple lead-based AI-enabled ECG algorithms on smartwatches for automated diagnosis of cardiac disorders. We also demonstrated the necessity of measuring at least 3 leads for accurate detection. Our results can be used as reference for the development of other AI models using sequentially measured asynchronous ECG leads via smartwatches for detecting various cardiac disorders.

scholarly journals THE RELATIONSHIP BETWEEN CALPROTECTIN AND PARAMETERS OF LIPID AND CARBOHYDRATE METABOLISM, ANTHROPOMETRIC AND INFLAMMATORY PARAMETERS IN PATIENTS WITH ACUTE MYOCARDIAL INFARCTION AND DIABETES MELLITUS TYPE 2

2019 ◽  
Vol 19 (2) ◽  
pp. 16-18
Author(s):  
M. I. Zhuravlova
Keyword(s):  
Diabetes Mellitus ◽  
Myocardial Infarction ◽  
Insulin Resistance ◽  
Acute Myocardial Infarction ◽  
Carbohydrate Metabolism ◽  
Diabetes Mellitus Type 2 ◽  
The Body ◽  
Diabetes Mellitus Type ◽  
The Relationship

Nowadays, an acute myocardial infarction is one of the leading causes of mortality among the population. The EHS-DH registry data clearly illustrate the association between the comorbidities and high mortality following acute myocardial infarction during a year period of follow up. The pronounced influence of carbohydrate metabolism disturbances on the survival of such patients has already been reported. The aim of the study was to analyze the immune inflammation relationships based on assessing calprotectin and the parameters of lipid and carbohydrate metabolism, to evaluate the presence and nature of the relationship between these parameters and carbohydrate metabolism parameters based on the study of blood glucose, insulin and insulin resistance (by the indices HOMA, QUICKI, Caro), anthropometric indicators and inflammatory indicators (monocyte and neutrophile levels). Materials and methods. The study included 64 patients (mean age 65, 31 ± 1.62 years) with acute myocardial infarction and concomitant diabetes mellitus type 2. The design of the study included the primary laboratory investigation of patients during the first day since the onset of acute myocardial infarction with the elevation of the ST segment before the initiation of thrombolytic therapy or percutaneous intervention. The direct correlation between the calprotectin concentration and the HOMA insulin resistance index (R = 0.52; p <0.05), insulinemia (R = 0.57; p <0.05), fasting glycaemia (R = 0, 59; p <0.05), as well as inverse correlation relationships between the Caro index (R = 0.68; p <0.05) and the QUICKI index (R = 0.59; p <0.05) were found out. Moreover, a direct correlation between calprotectin and triglyceride levels (R = 0.31; p <0.05), and negative correlation with high density lipoprotein (R = 0.35; p <0.05) was established as well. The level of total cholesterol and low density lipoproteins showed no significant association with the proinflammatory factor (R = 0.12; p> 0.05 and R = 0.18; p> 0.05, respectively). Conclusions. The increase in the body mass index and the activity of serum monocytes and neutrophils is associated with high concentrations of calprotectin that is accompanied by disturbances of carbohydrate homeostasis towards the growth of insulin resistance and changes of lipidograms of proatherrogenic nature.

scholarly journals IL-10–producing B cells are enriched in murine pericardial adipose tissues and ameliorate the outcome of acute myocardial infarction

2019 ◽  
Vol 116 (43) ◽  
pp. 21673-21684 ◽  
Author(s):  
Lan Wu ◽  
Rajeev Dalal ◽  
Connie D. Cao ◽  
J. Luke Postoak ◽  
Guan Yang ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
B Cells ◽  
Inflammatory Response ◽  
Healing Process ◽  
The Body ◽  
Adipose Tissues ◽  
Acute Mi ◽  
Antiinflammatory Cytokine ◽  
Specific Deletion

Acute myocardial infarction (MI) provokes an inflammatory response in the heart that removes damaged tissues to facilitate tissue repair/regeneration. However, overactive and prolonged inflammation compromises healing, which may be counteracted by antiinflammatory mechanisms. A key regulatory factor in an inflammatory response is the antiinflammatory cytokine IL-10, which can be produced by a number of immune cells, including subsets of B lymphocytes. Here, we investigated IL-10–producing B cells in pericardial adipose tissues (PATs) and their role in the healing process following acute MI in mice. We found that IL-10–producing B cells were enriched in PATs compared to other adipose depots throughout the body, with the majority of them bearing a surface phenotype consistent with CD5+ B-1a cells (CD5+ B cells). These cells were detected early in life, maintained a steady presence during adulthood, and resided in fat-associated lymphoid clusters. The cytokine IL-33 and the chemokine CXCL13 were preferentially expressed in PATs and contributed to the enrichment of IL-10–producing CD5+ B cells. Following acute MI, the pool of CD5+ B cells was expanded in PATs. These cells accumulated in the infarcted heart during the resolution of MI-induced inflammation. B cell-specific deletion of IL-10 worsened cardiac function, exacerbated myocardial injury, and delayed resolution of inflammation following acute MI. These results revealed enrichment of IL-10–producing B cells in PATs and a significant contribution of these cells to the antiinflammatory processes that terminate MI-induced inflammation. Together, these findings have identified IL-10–producing B cells as therapeutic targets to improve the outcome of MI.

scholarly journals Prediction and Prognosis of Acute Myocardial Infarction in Patients with Previous Coronary Artery Bypass Grafting Using Neural Network Model (Preprint)

2019 ◽  
Author(s):  
Predrag Mitrovic ◽  
Branislav Stefanovic ◽  
Mina Radovanovic ◽  
Nebojsa Radovanovic ◽  
Dubravka Rajic ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
Artificial Neural Network ◽  
Coronary Artery ◽  
Coronary Artery Bypass ◽  
Artery Bypass ◽  
Bypass Grafting ◽  
Data Set

BACKGROUND Patients with previous coronary artery bypass grafting represent a substantial percentage of the total population of patients with acute myocardial infarction. Prognosis of the future disease expression is an important part in the follow-up of patients with previous CABG. It is well known that outcome of patients with previous CABG influenced with a lot of abnormalities. Neural networks are a form of artificial intelligence and they may obviate some of the problems associated with traditional statistical techniques, and they are representing a major advance in predictive modeling. OBJECTIVE The purpose of this study was to assess the usefulness and accuracy of artificial neural network in the prediction and prognosis of acute myocardial infarction in patients with previous coronary artery bypass surgery. METHODS The baseline characteristics and clinical data were recorded in 2180 consecutive patients. The data set contains 13 predictor variables per patient. It was first randomly split into training (1090 cases) and test sets (1090 cases). Artificial neural network performance was evaluated using the original data set for each network, as well as its complementary test data set, containing patient data not used for training the network. The program compared actual with predict outcome for each patient, generating a file of comparative results. At the end, results from this file were analyzed and compared, on the basis of a 2x2 contingency table constructed from expected or obtained statistics (accuracy, sensitivity, specificity and positive/negative predictivity). RESULTS Linear discriminant analysis was not efficient for prediction and prognosis of acute myocardial infarction in patients with prior CABG. The results show that a statistical linear model is not able to perform class separation in multidimensional space and that a nonlinear approach is justified. In analyzing the performance of neural network in outcome prognosis of AMI in patients with previous CABG it is clear that neural network method was better for almost all statistic parameters for all analyzed prediction variables. CONCLUSIONS In this clinical situation, artificial intelligence appears to be superior to linear methods for prediction and prognosis of AMI in patients with previous CABG.

Pharmacoepidemiology of Bleeding Events after Use of R-Alteplase or Streptokinase in Acute Myocardial Infarction

1993 ◽  
Vol 27 (7-8) ◽  
pp. 956-962 ◽  
Author(s):  
Donald C. McLeod ◽  
W. Gerald Coln ◽  
Charles F. Thayer ◽  
Eleanor M. Perfetto ◽  
Abraham G. Hartzema
Keyword(s):  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
Logistic Regression ◽  
Medical Record ◽  
Recombinant Tissue Plasminogen Activator ◽  
The Body ◽  
Bleeding Complications ◽  
Bleeding Events ◽  
Clinically Significant ◽  
Concomitant Medications

OBJECTIVE: To determine in nonresearch, general medical practice conditions the comparative incidence and types of bleeding complications after the use of streptokinase (SK) and r-alteplase (recombinant tissue plasminogen activator, rt-PA) to treat acute myocardial infarction (AMI). DESIGN: Retrospective medical record review of concurrently treated patients (96-hour observation posttreatment) in 32 participating hospitals in the US. MAIN OUTCOME MEASURES: The medical record description of all bleeding events regarding the body site affected, changes in hemoglobin concentrations, blood products administered, and clinical outcome (permanent sequelae or death). Bleeding severity was determined by defined criteria. CONTROL DATA: Comorbidity and concomitant medications (e.g., aspirin, heparin, warfarin) likely to predispose or contribute to bleeding events were analyzed. DATA ANALYSIS: Logistic regression analysis. RESULTS: Data from 419 patients who received rt-PA and 207 who received SK were evaluated. In the 96-hour period after initiation of thrombolytic therapy, 30.5 and 31.9 percent of rt-PA and SK patients, respectively, experienced one or more bleeding events (crude risk ratio [CRR] = 1.04; 95 percent confidence interval [CI] 0.91–1.14; p=0.73). In the first 24-hour period, 21.5 percent of rt-PA and 15.9 percent of SK patients experienced bleeding events (CRR = 0.74; 95 percent CI 0.42–1.15; p=0.08). The leading types of bleeding and percents of all patients affected were: Perivascular access site (18.4 percent), gastrointestinal (6.4 percent), skin/soft tissue/muscle (5.0 percent), urinary (3.4 percent), pulmonary (2.2 percent), systemic (1.9 percent), and oral (1.4 percent). Intracranial bleeding occurred in 4 rt-PA and 2 SK patients; 4 of these patients died. Events deemed clinically significant occurred in 15 rt-PA and 9 SK patients (3.8 percent of all patients). Ten patients likely died from these events, 6 within the first 24 hours. Three rt-PA patients and 1 who received SK (0.6 percent) died of cerebrovascular events within the first 24 hours. After controlling for demographic factors and therapeutic variables, using logistic regression analyses, no thrombolytic-related differences were found in the incidence or severity of bleeding following use of the two thrombolytics. CONCLUSIONS: These clinical data do not support a theoretical advantage of rt-PA to cause less bleeding propensity than SK.

Sign in / Sign up

Export Citation Format

Share Document