Geri Dön

Otomatik aritmi dedeksiyonu

Başlık çevirisi mevcut değil.

  1. Tez No: 39096
  2. Yazar: GÜNNUR ALANYALI
  3. Danışmanlar: DOÇ.DR. MEHMET KORÜREK
  4. Tez Türü: Yüksek Lisans
  5. Konular: Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
  6. Anahtar Kelimeler: Belirtilmemiş.
  7. Yıl: 1993
  8. Dil: Türkçe
  9. Üniversite: İstanbul Teknik Üniversitesi
  10. Enstitü: Fen Bilimleri Enstitüsü
  11. Ana Bilim Dalı: Belirtilmemiş.
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 82

Özet

ÖZET Bu tezde otomatik: Elektrokardiografik CEKGD aritmi dedeksiyonu yapabilmek için gerekli adımlar incelenecektir. EKG işaretinin oluşumu, normal EKG işareti, EKG işaretindeki bozukluklar, EKG işaretinin algılanması konularının ardından bu işaretin bilgisayar yardımı ile incelenmesini sağlayacak olan algoritmalar incelenecektir. EKG işaretindeki ritm ve şekil bozuklukları incelenecek ve bu bozuklukları bilgisayar yardımı ile bulabilecek programların yazılmasına çalışılacaktır. EKG işaretindeki ritm ve şekil bozukluklarına aritmi adı verilir. Aritmi hastaları sürekli gözetim altında tutulan ve yoğun hemşire bakımı isteyen hastalardır. Bu yüzden hastalara bağlanan aritmi dedektarleri sayesinde yoğun bakım ünitelerinde bir rahatlama görülmüştür. Hastanın sürekli olarak incelenmesi sırasında EKG monitörleri ve alarm veren aritmi dedektarleri kullanılmaktadır. Böylece hastanın gunluk kayıtları tutulmakta ve ani ol Um riski azaltılabiimektedir. Günümüzde tam anlamıyla mükemmel olmayan bilgisa2var teşhisleri uzman bir doktor tarafından gözden geçirilerek teşhise varılmaktadır. Gene de bilgisayar tarafından yapılan EKG incelemeleri fiyatları düşürmekte ve daha geniş kullanıma açmaktadır.

Özet (Çeviri)

SUMMARY AUTOMATED ARRHYTHMIA DETECTION The electrocardiogram CECGD is the recording of the electric activity of the heart. The mechanical activity of the heart function is linked with the electrical activity. The ECG therefore is an important diagnostic tool for assessing heart function. The heart electric cycle begins at si no atria CSAD node on the right atrium; it is an astable bundle of nerves. The SA node paces the heart. SA node impulses cause the contraction of the atria generating the P wave in the ECG. The impulses travel along conduction fibers within the atria to the atria ventricular CAvO node which controls impulse transmission between atria and ventricles. A special conduction system, consisting of the bundles of His and Purkinje, transfers the impulses into the lower and outer parts of the ventricles. The contraction of ventricles comprises the pumping action of the heart and generates the QRS complex in the ECG. About ISO msec later the ventricles repolar ize, causing the T wave in the ECG. The repolarization of the atria is rarely seen in the ECG. In the rare cases in which it does occur, it appears between the P and Q waves and is called TA wave. An additional wave, the U wave, is sometimes recorded after the T wave. Its cause is believed to be the repolarization of ventricular muscl es.The cardiac rhythm is a random process. It is usually- measured by the RR interval. If the heart rate slows down it calls bradycardia. It accelerates in tachycardia. Rhythm disturbances, arrhythmia, may arise under several abnormal conditions. Sometimes a portion of the myocardium discharges independently, causing a heart beat not in the normal SA sequence, this is known as actopic beat extr asystole or preventricular contraction CPVO. When independent discarges continue, the heart may enter a state of atrial or ventricular fibrillations. Conventional ECG consists of the PQRST complex with amplitudes of several milivolts. it is usually processed in the frequency band of 0. OS to lOO Hz where most most of the energy of the ECG is included. The first step in ECG processing is the identification of the R wave. This is done in order to synchronize consecutive complexes and for RR interval analysis. Various techniques of wavelet detection have been employed, the problem is particularly severe when recording the ECG under active conditions where muscle signals and other noise sources obscure the QRS complex. The analysis of the RR interval is an important part of heart patient monitoring. Analog filtering helps to eliminate unwanted frequency components in the ECG signal. Noise problems don't terminate because baseline wander, movement artifact, and muscle noise have components that lie within the same frequency band as the ECG. This type of noise can't be completely filtered out. VIDigital filters are si mi liar to analog filters but allow much more precise control over their characteristics. They have the disadvantage that the more effective they are, the more computer time they require. A much more universal technique is called eye-closing, which is commonly applied to the muscle-noise problem where the unwanted components lie in the high end of the ECG frequency band. The computer monitors the amout of high frequency components in the incoming ECG signal by counting zero- crossings or the number of upward and downward slopes per unit time. If the number of these events exceeds a threshold, a message is issued to the subsequent processing stages to suspend analysis for a short time. During this time the noise analyzer continues to monitor the amount of noise in the signal. Movement artifact is the most troublesome noise for arrhythmia detectors, since it is composed predominately of low frequencies and often mimics the shape of PVCs. Two approaches to this problem involve the use of an independent measure of electrode motion. The first makes use of a second set of electrodes which are placed exactly the same recording positions as the original set. The two sets of electrodes record from same sites, the detected physiologic signals should be identical on both leads. Any difference in potential between the two leads must be due to motion artifact at one or both electrode sites. This smart electrode tecnique shows to reduce false alarms substantially in clinical monitoring settings. VI iA second approach to electrode motion artifact detection relies on the measurement of the electrode/skin impedance. Motion of the electrode with respect to the skin causes changes in the electrode/skin impedance as well as producing the electrical artifact. QRS detection is the first analysis stage in which pattern recognition is important. The most general QRS criterion is that the waveform have a certain minimum amplitude, the purpose of which eliminate to a first approximation the false detection of P waves, T waves, and low to medium level noise. This amplitude threshold may be fixed or may vary depending on the height of previously detected compl exes. Correlation is a technique often used for shape characterization. This method compares each new QRS waveform with previously detected QRS shapes. A single normal waveform or template is stored. The comparison is performed by the correlation coefficent of the stored normal QRS with the unknown QRS. A fixed number of samples CND is used in the calculation. These N samples are centered on the previously determined fiducial point for both the model and for the new QRS. The time of occurrence of the new QRS and its correlation with each stored model or template is then passed to the next stage for classification. The shape characterization process varies from the single- stored normal form to the very complex multi -dimensional feature space. Shape characterization, however it is performed, VI 11sets the stage for beat classification. The basic information needed for beat classification is the average RR interval and the characteristics of the normal beat. The final stage in arrhythmia detection is the classification of rhythms or defined sequences of beats. Rhythm diagnosis can be accomplished using only RR interval i nf or ma t i on. The important arrhythmias CPVC, atrial or ventricular f ibrillation3 must be detected and the doctors must see them. ECG signal is one of the most important vital signs monitored from cardiac patients. ECG recording, analysis and continuous monitoring may take place in a doctor's office or a cardiac intensive care unit in a hospital. Cardiologists readly interpret the ECG waveforms and classify them in normal and abnormal patterns. In giving their interpretations, the cardiologists take into considaration information such as patient history, heart rate shapes of QRS complex and other more subtle features. Interpretations by the practiced eye of a cardiologist baced by clinical knowledge and experience are indispensable in providing a complete diagnostic. Computers can assist a cardiologist in the task of ECG monitoring and interpretations. For example, in a CICU C Coronary Intensive Care Unit}, ECGs of several patients must be monitored continuously to detect any life t hr eating abnormalities that may occur. Computerized ECG monitors that provided complete 12-lead diagnostic-quali ty ECG recordings and interpretations have become common. IXThis thesis anal i ses ECG signal with computer and detects the arrhythmias like tachycardia, bradicardia, PVC using C programming language. Also it analises other feature arrhythmias with correlation coefficient, city block, straigth line and matched filtering using C programming language. x

Benzer Tezler

  1. Tez No

    896397

    Automatic arrhythmia classification from electrocardiogram measurements with deep learning

    Derin öğrenme ile elektrokardiyogram ölçümlerinden otomatik aritmi sınıflandırma

    BERKCAN YURTSEVER

    Yüksek Lisans

    İngilizce

    İngilizce

    2024

    Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolHacettepe Üniversitesi

    Bilgisayar Mühendisliği Ana Bilim Dalı

    PROF. DR. EBRU AKÇAPINAR SEZER

  2. Tez No

    255847

    Classification of ECGarrythmia beats with artificial neural networks

    Yapay sinir ağlarıyla EKG aritmi sınıflandırılması

    SEÇİL ZEYBEKOĞLU

    Yüksek Lisans

    İngilizce

    İngilizce

    2009

    BiyomühendislikBoğaziçi Üniversitesi

    Biyomedikal Mühendisliği Ana Bilim Dalı

    PROF. DR. MEHMED ÖZKAN

  3. Tez No

    283699

    Multi-stage classification of abnormal patterns in EEG and e-ECG using model-free methods

    Modelden bağımsız yöntemler kullanılarak EEG ve EKG içindeki anormal örüntülerin çok katlı sınıflandırılması

    YAKUP KUTLU

    Doktora

    İngilizce

    İngilizce

    2010

    Elektrik ve Elektronik MühendisliğiDokuz Eylül Üniversitesi

    Elektrik-Elektronik Mühendisliği Ana Bilim Dalı

    YRD. DOÇ. DR. DAMLA KUNTALP

  4. Tez No

    276763

    Erken ventriküler kasılmalarda ysa tabanlı bir sınıflandırıcının FPGA ile gerçekleştirilmesi

    An FPGA implementation of ann-based premature ventricular contraction classifier

    AHMET TURAN ÖZDEMİR

    Doktora

    Türkçe

    Türkçe

    2010

    Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolErciyes Üniversitesi

    Elektronik Mühendisliği Ana Bilim Dalı

    PROF. DR. KENAN DANIŞMAN

  5. Tez No

    903754

    Convolutional autoencoder based heart arrhythmia detection system

    Evrişimli otomatik kodlayıcı tabanlı aritmi tespit sistemi

    ÖYKÜ ERAVCI

    Yüksek Lisans

    İngilizce

    İngilizce

    2024

    Elektrik ve Elektronik MühendisliğiYaşar Üniversitesi

    Elektrik-Elektronik Mühendisliği Ana Bilim Dalı

    DR. ÖĞR. ÜYESİ NALAN ÖZKURT

Geri Dön