Detection and diagnosis of myocarditis in young patients using ECG analysis based on artificial neural networks
- 95 Downloads
Myocarditis is a common disease of the pediatric circulatory system, and its clinical manifestations are diversified. Especially in older patients, chest tightness, shortness of breath and palpitations may occur. However, some patients may have no distinct clinical symptoms and may occasionally experience symptoms during the diagnosis and treatment of respiratory infections, such as arrhythmias in infants and young patients. Because the clinical manifestations of myocarditis are not typical, the specificity and sensitivity of serum myocardial enzyme detection are not very high, which hinders a definitive diagnosis of certain cases by clinicians. However, degeneration, necrosis and inflammatory diseases of cardiomyocytes can affect myocardial conductivity and stress. In patients with myocarditis, an electrocardiogram (ECG) signal changes abnormally; thus, the ECG remains an indispensable means of diagnosing myocarditis. Biomedical signal detection and signal processing techniques can facilitate the study of biological and physiological systems and assist physicians in the diagnosis and treatment of patients. However, traditional manual analysis of ECG methods has some drawbacks. Due to the extensive variety of electrocardiograms, variations in various waveform patterns are subtle and complex. Clinicians often need a solid theoretical foundation and a wealth of clinical experience to make a correct assessment using an ECG. If the medical expert is engaged in a large number of ECG recognition studies for a long time, fatigue may develop and cause misdiagnosis. In addition, the manual analysis of an ECG is part of the postmortem analysis, achieving a real-time diagnosis is difficult, and the resulting delay is likely to cause critical patients to miss the best treatment opportunity. To solve this problem, we have developed a dynamic electrocardiogram medical-assisted diagnosis system to assist medical experts in the diagnosis and treatment of patients with myocarditis. Our system can process the collected ECG signals, obtain waveform parameters, and automatically perform an analysis according to an expert’s clinical experience. These parameters provide diagnostic recommendations.
KeywordsECG signal Myocarditis Signal detection Signal processing Medical aided diagnosis system
Mathematics Subject Classification00-XX
This work was supported in part by the National Key R&D Program (No. 2018YFC0831006-3 and 2017YFB1400102), the Key Research and Development Plan of Shandong Province (No. 2017CXGC1503 and 2018GSF118228).
- 2.Poulakaki N, Papalois A, Papadimitriou D, Prachalias A, Stevens C, Biomed C, Papadimitriou L, John P (2001) First clinical application of a second generation HepatAssist bioartificial liver support system without charcoal associated with neurological recovery in a fulminant hepatic failure patient. Gastroenterology 120(5):A543–A543CrossRefGoogle Scholar
- 4.Nadi S, Madhuka J, Pham TT (2018) A low-complexity algorithm for detection of atrial fibrillation using an ECG. Physiol Meas 49(10):28–42Google Scholar
- 5.Lee JW, Lee GK (2014) Applying nonlinear dynamics to ECG signal processing. Int J Control Autom Syst 3(1):137–142Google Scholar
- 12.Schmidt M, Krug JW, Rose G (2016) Reducing of gradient induced artifacts on the ECG signal during MRI examinations using Wilcoxon filter. Curr Direct Biomed Eng 2(1):175–178Google Scholar
- 14.Atanassova P, Tokmakova M, Djurkova A, Naydenov V, Chalakova N, Dimitrov B (2006) Abnormal ECG patterns during the acute phase of subarachnoid hemorrhage in patients without previous heart disease. Cent Eur J Med 1(2):148–157Google Scholar
- 16.Schmidt M, Krug JK, Rose G (2016) Real-time QRS detection using integrated variance for ECG gated cardiac MRI. Curr Direct Biomed Eng 2(1):255–258Google Scholar
- 18.Wu Q, Ye J, Yang J (2016) Relation between carotid intima-media thickness and diastolic augmentation index in Chinese population. In: 9th International congress on image and signal processing on biomedical engineering and informatics (CISP-BMEI). IEEE, pp 1761–1765Google Scholar
- 23.Dassen WR, Schwab J, Rogg HJ, Pauschinger M, Fessele K, Bareiter T, Bär I, Loose R (2016) Functional and morphological parameters with tissue characterization of cardiovascular magnetic imaging in clinically verified “infarct-like myocarditis”. Fortschr Röntgenstr 188(04):365–373Google Scholar
- 28.Hu YH, Tompkins WJ, Urrusti JL (1993) Applications of artificial neural networks for ECG signal detection and classification. J Electrocardiol 26(Suppl 1):66–73Google Scholar