Arrhythmia Detection by Using Chaos Theory with Machine Learning Algorithms

Heart monitoring improves life quality.Electrocardiograms(ECGs or EKGs)detect heart irregularities.Machine learning algorithms can create a few ECG diagnosis processing methods.The first method uses raw ECG and time-series data.The second method classifies the ECG by patient experience.The third technique translates ECG impulses into Q waves,R waves and S waves(QRS)features using richer information.Because ECG signals vary naturally between humans and activities,we will combine the three feature selection methods to improve classification accuracy and diagnosis.Classifications using all three approaches have not been examined till now.Several researchers found that Machine Learning(ML)techniques can improve ECG classification.This study will compare popular machine learning techniques to evaluate ECG features.Four algorithms—Support Vector Machine(SVM),Decision Tree,Naive Bayes,and Neural Network—compare categorization results.SVM plus prior knowledge has the highest accuracy(99%)of the four ML methods.QRS characteristics failed to identify signals without chaos theory.With 99.8%classification accuracy,the Decision Tree technique outperformed all previous experiments.

Wide QRS complex tachycardia in a patient with wide QRS complex sinus rhythm due to left bundle branch block pattern

A 58-year-old man with hypertension and half a year ago documented left bundle branch block(LBBB)in his past history presented with progressive dyspnea.A pre-admission echocardiography showed severely depressed systolic left ventricular function[ejection fraction(EF)=27%],diffuse hypokinesis and dilated heart chambers consistent with dilated cardiomyopathy(DCM).

Short-term Prognosis of Fragmented QRS Complex in Patients with Non-ST Elevated Acute Myocardial Infarction

Background：There remains significant debate as to the relationship between fragmented QRS （fQRS） complexes on electrocardiogram （ECG） and acute myocardial infarction （AMI）.Few studies have reported on this relationship in non-ST elevated AMI （NSTEMI）,and thus,we attempt to assess this relationship and its potential short-term prognostic value.Methods：This was a single-center,observational,retrospective cohort study.A total of 513 consecutive patients （399 men,114 women） with NSTEMI within 24 h who underwent coronary angiography at our department,between January 1,2014,and December 31,2014.Patients were divided into 2 groups according to the presence or absence of fQRS complex on the admission ECG.fQRS complexes were defined as the existence of an additional R＇ or crochetage wave,notching in the nadir of the S wave,RS fragmentation,or QS complexes on 2 contiguous leads.All patients were followed up for 6 months,and all major adverse cardiac events （MACE） were recorded.Results：In this study,there were 285 patients with fQRS ECG in the 513 patients with NSTEMI.The number of patients with 0-2 coronary arteries narrowed by ≥50% in fQRS group were less while patients with 3 narrowed arteries were more than in the non-fQRS group （P =0.042）.There were fewer Killip Class Ⅰ patients in the fQRS group （P =0.019）,while Killip Class Ⅱ,Ⅲ,and Ⅳ patients were more in the fQRS group than in the non-fQRS group （P =0.019）.Left ventricular ejection fraction levels were significantly lower in the fQRS group （P =0.021）.Baseline total cholesterol,low-density lipoprotein,creatinine,creatine kinase,homocysteine,high-sensitivity C-reactive protein （CRP）,and red blood cells distribution width levels were significantly higher in the fQRS group.Total MACE （MACE,P =0.028）,revascularization （P =0.005）,and recurrent angina （P =0.005） were also significantly greater in the fQRS group.On final logistic regression analysis,after adjusting for baseline variables,the following variables were independent predictors of fQRS：Coronary artery narrowing （P =0.035）,Killip classification （P =0.026）,and total cholesterol （P =0.002）.The following variables were found to be independent predictors of preoperative MACE：Hemoglobin （P =0.000）,gender （P =0.026）,fQRS （P =0.016）,and time from myocardial infarction to balloon or coronary artery bypasses grafting （P =0.013）.Conclusions：The fQRS complexes are commonly present in NSTEMI and the fQRS complexes are an independent predictor of MACE in NSTEMI patients.The number of narrowed coronary arteries,Killip classification,and total cholesterol are all independent predictors of the fQRS complexes.

Efficacy of cardiac resynchronized therapy on patients with a fragmented QRS complex

Background Based on previous studies, a fragmented QRS （fQRS） complex, as a predictive biomarker of myocardial scarring condition, could be used to predict the outcomes of cardiac resynchronized therapy （CRT）. However, this conclusion is still debatable. Methods Fifty ischemic or non-ischemic cardiomyopathy patients failure （aged 65 ± 10 yrs, 34 males, 16 females） with refractory heart, diagnosed by the criteria of New York Heart Association received CRT. The ECGs of 18 patients with a fQRS complex （divided by Das） were compared with those of 32 patients without a fQRS complex, who were evaluated by 12-lead ECG before CRT. The patients were followed up for six months, and 12-lead ECG and echocardiography were reviewed. At least 15% reduction in the left ventricular end-systolic volume （LVESV） was defined as responders according to the data obtained for between-group and intra-group analysis. Results Six patients （33.3%） in the fQRS group and 24 patients （75%） in the non-fQRS group responded well. In addition, comparisons of indicators from surface ECG and echocardiography 6 months after CRT showed that the non-fQRS group benefited from CRT significantly more than the fQRS group. Conclusions The fQRS complex has good predictive value for responsiveness to CRT. Non-fQRS complex patients with refractory heart failure may benefit more from CRT, and these patients need to receive this treatment as early as possible.