• Journal of information and communication convergence engineering
• /
• 제19권2호
• /
• pp.93-101
• /
• 2021

Cardiac arrhythmias are common heart diseases and generally cause sudden cardiac death. Electrocardiogram (ECG) is an effective tool that can reveal the electrical activity of the heart and diagnose cardiac arrhythmias. We propose detection of P waves based on QRST cancellation zero-one substitution. After preprocessing, the QRST segment is determined by detecting the Q wave start point and T wave end point separately. The Q wave start point is detected by digital analyses of the QRS complex width, and the T wave end point is detected by computation of an indicator related to the area covered by the T wave curve. Then, we determine whether the sampled value of the signal is in the interval of the QRST segment and substitute zero or one for the value to cancel the QRST segment. Finally, the maximum amplitude is selected as the peak of the P wave in each RR interval of the residual signal. The average detection rate for the QT database was 97.67%.

• 대한수의학회지
• /
• 제33권4호
• /
• pp.719-734
• /
• 1993

The electrocardiographic(ECG) parameters on the standard limb leads in the normal Korean native cattle have been measured with a 3 channel Electrocardiograph built in a computed analysis. The study was conducted on the animals 98 heads of mean age of 17.7 months. Conduction parameters, waves, intervals and segments have been recorded. The recordings were analyzed as to shape and amplitude of the P and T waves and the components of the QRS complex. Heart rate was recorded by the Electrocardiogram which were a mean of $80.4{\pm}11.6beats/min$. And the younger had a higher heart rate than the older one. Average conduction times in the RP, the QRS complex and the QTc interval recorded $166.7{\pm}23.1msec.$, $79.7{\pm}8.8msec.$ and $395.5{\pm}30.4msec.$, in the P and T wave duration recorded $70.1{\pm}13.5msec.$ and $97.6{\pm}16.9msec.$, and in the PR and ST segment duration recorded $97.9{\pm}23.5msec.$ and $173.9{\pm}40.3msec.$, respectively. The wave forms in each lead observed various types. The amplitudes of wave type showed the highest frequency in each lead that were analyzed as follow : 1. In P wave, amplitudes of the positive type showed the frequency of 65.3%, 82.7% and 52.0% in leads I, II and III that were $103.1{\pm}47.8{\mu}V$, $115.2{\pm}37.3{\mu}V$ and $67.4{\pm}26.9{\mu}V$, and it showed the frequency of 54.1% and 85.7% in the leads aVL and aVF that were $63.7{\pm}23.0{\mu}V$, $88.0{\pm}83.6{\mu}V$, respectively. Average amplitude of the negative type showed the frequency of 78.6% in lead aVR which was $99.3{\pm}38.0{\mu}V$. 2. Average amplitude of the QRS complex were from $362.8{\pm}177.7{\mu}V$ to $532.8{\pm}253.9{\mu}V$(mean of $449.1{\pm}57.2{\mu}V$) that in all leads except lead I were manifested the Low-Voltage QRS complex(below 0.5mV). Average amplitudes of each wave type in the QRS complex aere $-50.2.4{\pm}258.2{\mu}V$ and $-428.6{\pm}195.1{\mu}V$ in the QS groups type that showed a frequency of 66.3%, 70.4% in the leads I and aVL, were $451.1{\pm}20.4.0{\mu}V$, $387.6{\pm}175.8{\mu}V$ and $299.3{\pm}146.5{\mu}V$ in the R groups type that showed a frequency of 48.0%, 53.1% and 34.7% in the leads III, aVR and aVF, and were $-307.5{\pm}180.3{\mu}V$, $201.4{\pm}77.2{\mu}V$ in the QR wave type which showed a frequency of 39.8% in lead II, respectively. 3. In T wave, amplitude of the positive type showed the frequency of 50.0%, 82.7%, 51.0% and 57.1% in leads II, III aVR and aVF which were $214.9{\pm}115.6{\mu}V$, $188.5{\pm}119.3{\mu}V$, $191.0{\pm}93.7{\mu}V$ and $165.7{\pm}91.9{\mu}V$, and the negative type showed a frequecny of 66.3% and 72.5% in leads I and aVL. that were $221.3{\pm}112.5{\mu}V$, $-173.6{\pm}86.7{\mu}V$, respectively. 4. Amplitude of ST segment in leads I, II and III were a mean of $-12.2{\pm}37.2{\mu}V$, $17.5{\pm}42.6{\mu}V$ and $28.3{\pm}40.4{\mu}V$, in leads aVR, aVL and aVF were $-3.9{\pm}32.5{\mu}V$, $-15.9{\pm}35.6{\mu}V$ and $26.2{\pm}37.5{\mu}V$, respectively.

• 한국정보통신학회논문지
• /
• 제19권1호
• /
• pp.192-200
• /
• 2015

부정맥 분류를 위한 기존 연구들은 특정 ECG 데이터에 종속적으로 개발되었기 때문에 다른 환경에 적용할 경우 그 성능에 변화가 많아 임상 적용에 한계가 있다. 즉, 생체 신호의 특성상 개인 간의 차이가 있음에도 불구하고, 일반적인 ECG 신호의 판단규칙에 따라 진단을 수행하기 때문이다. 또한 이러한 대부분의 방법들은 P, Q, R, S, T 지점의 정확한 측정을 필요로 하며, 데이터의 가공 및 연산이 복잡하다. 따라서 이러한 문제점을 극복하기 위해서는 개인별 특성을 가진 ECG 데이터를 분석하여 최소한의 특징점을 추출함으로써 그에 따른 패턴을 분류하는 것이 필요하다. 본 연구에서는 이상 심전도와 같은 다양한 신호를 고려하여 Q, R, S 피크 변화에 따른 개인별 ECG 신호의 패턴 분석기법을 제안한다. 이를 위해 전처리를 통해 잡음이 제거된 심전도 신호에서 R파를 검출하고 Q, R, S의 진폭과 위상변화에 따른 8개의 특징점을 추출하였다. 이후 각 특징점의 피크 변화와 형태에 따른 ECG 신호를 분석하고 부정맥 유형에 따른 9가지 패턴을 정의하였다. 제안한 방법의 우수성을 입증하기 위해 43개의 MIT-BIH 레코드를 대상으로 Normal, PVC, PAC, LBBB, RBBB, Paced Beat의 각 패턴을 분석하였다. 실험결과 9가지 패턴에 대한 검출율은 93.72%로 우수하게 나타났다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2002년도 가을 학술발표논문집 Vol.29 No.2 (1)
• /
• pp.199-201
• /
• 2002

본 논문에서는 심장질환의 진단 알고리즘의 개발에 있어서 필수적으로 요구되는 심장질환별 ECG 데이터의 수집에 관하여 기술한다. 또한, 진단 알고리즘을 개발하기 위한 전단계로서 심전도 신호에서 각 특징들을 검출하는 알고리즘에 관하여 설명하고, 이를 MITDB와 수집한 ECG 신호에 적용한 결과를 보인다. QRS-complex의 검출은 99% 이상의 정확도를 보이나, P-wave와 T-wave의 검출에서는 아직까지 보완할 점이 많은 것으로 나타난다. 심장질환별 12-채널 ECG 데이터베이스의 구축은 보다 정확하고 현실적인 진단 알고리즘을 개발하는 데 크게 기여할 것으로 기대한다.

• 한국임상수의학회지
• /
• 제19권2호
• /
• pp.132-138
• /
• 2002

Non-human primates are widely used for experimental animal and raised as companion animal in Korea. To establish the electrocadiogram (ECG) of non-human primates that are domestically raised, the author measured bipolar limb leads and augmented unipolar limb leads, after tiletamine/zolazepam (TZ) injection as sedative agents. The amplitudes of P,Q, R,S and T wave and duration time of P wave, QRS complex, PR and QT interval in each lead of ECG were evaluated in 7 non-human primates at 15 minutes after TZ injection, respectively. The amplitude of P wave in I, II,III, aVR, aVL and aVF leads revealed 0.06$\pm$ 0.05,0.14$\pm$ 0.05, 0.1 $\pm$ 0.05,-0.11 $\pm$ 0.06,-0.04$\pm$ 0.04 and 0.12$\pm$ 0.05 mV respectively. The amplitude of Q wave revealed -0.16$\pm$ 0.15, -0.23$\pm$ 0.18, -0.17$\pm$ 0.13, 0.16$\pm$0.13, 0.04$\pm$ 0.09 and -0.2$\pm$0.13 mV, respectively. The amplitude of R wave revealed 0.56$\pm$0.56, 1.24$\pm$ 0.67, 0.92$\pm$0.33, -0.37$\pm$ 1.14, -0.22$\pm$ 0.47 and 1.12 $\pm$ 0.47 mV, respectively. The amplitude of S wave revealed -0.02$\pm$ 0.08, -0.04$\pm$0.06, -0.06$\pm$0.04, 0.02$\pm$0.04, 0.04$\pm$0.09 and -0.04 $\pm$ 0.06 mV, respectively. The amplitude of T wave revealed -0.01 $\pm$ 0.15,-0.02$\pm$ 0.13, 0.01 $\pm$ 0.08, 0.02$\pm$ 0.12, 0.01 $\pm$ 0.11 and -0.03$\pm$ 0.09 mV, respectively. The duration time of P wave revealed 0.05 $\pm$ 0.01, 0.04$\pm$ 0.01, 0.05$\pm$ 0.02, 0.05 $\pm$ 0.02, 0.04$\pm$ 0.01 and 0.04$\pm$ 0.01 sec, respectively. The duration time of QRS complex revealed 0.05 $\pm$ 0.02,0.05$\pm$ 0.01, 0.05 $\pm$ 0.01, 0.04$\pm$ 0.01, 0.05$\pm$ 0.01 and 0.05 $\pm$ 0.01 sec, respectively. The duration time of PR interval revealed 0.08$\pm$ 0.01, 0.07$\pm$0.01,0.08$\pm$ 0.03, 0.08$\pm$0.01, 0.08$\pm$ 0.01 and 0.08$\pm$0.01 sec, respectively. The duration time of QT interval revealed 0.23$\pm$ 0.06, 0.22$\pm$ 0.05, 0.23 $\pm$ 0.06, 0.23$\pm$ 0.06, 0.24$\pm$ 0.05 and 0.22$\pm$ 0.02 sec, respectively. No significant changes were observed in e amplitude of P and T waves. The amplitude of QRS complex in ketamine group was higher than that of TZ group. However, no significant changes were observed in both intra-group and inter-group. There were no significant changes in the duration time of P wave, QRS complex and PR interval obtained from both groups. Also, the duration time of QT interval in TZ group was significantly longer at 30 min.(P< 0.05) an that of 10 minutes after injection. However, significant difference was not found between two groups. The mean cardiac electric axis in ketamine group tended to decrease until 30 min. after injection and then gradually increase. However, mean cardiac electric axis of TZ group was increased until 30 min. after injection and then decreased. But significant differences were not observed between groups. These results suggest that the ECG pattern after TZ injection to non-human primates reared in korea was established. It was also considered that the injection of ketamine and TZ didn't significantly affect on ECG pattern of the non-human primates.

• 한국정보통신학회논문지
• /
• 제23권2호
• /
• pp.117-126
• /
• 2019

부정맥 분류를 위한 기존 연구들은 분류의 정확성을 높이기 위해 신경망, 퍼지, 시계열 주파수 분석, 비선형 분석법 등이 연구되어 왔다. 이러한 방법들은 분류율를 향상시키기 위해 정확한 특징점과 많은 양의 신호를 처리해야 하기 때문에 데이터의 가공 및 연산이 복잡하며, 다양한 부정맥을 분류하는데 어려움이 있다. 본 연구에서는 AR(Auto Regressive) 모델링 기반의 특징점 추출과 SVM(Support Vector Machine)을 통한 조기수축 부정맥 분류 방법을 제안한다. 이를 위해 잡음을 제거한 ECG 신호에서 R파를 검출하고 QRS와 RR 간격의 특정 파형 구간을 모델링하였다. 이후 최적 세그먼트 길이(n1, n2), 최적 차수( p1, p2)의 4가지 AR 모델링 변수를 추출하고 SVM을 통해 Normal, PVC, PAC를 분류하였다. 연구의 타당성을 입증하기 위해 MIT-BIH 부정맥 데이터베이스를 대상으로 한 R파의 평균 검출 성능은 99.77%, Normal, PVC, PAC 부정맥은 각각 99.23%, 97.28, 96.62의 평균 분류율을 나타내었다.

• 한국임상수의학회지
• /
• 제14권2호
• /
• pp.319-337
• /
• 1997

The electrocardiographic (ECG) parameters in the normal Korean black goat have been measured with a 3-channel Electrocardiograph and computed, analyzed. ECG in 243 black goats were made with the limb leads (I, II, III, aVR, aVL, aVF), that were recorded conduction parameters of wave and interval, and were analyzed as to shape and amplitude of the P and T waves and the components of the QRS complex. Heart rate were recorded by the ECG which were a mean of 106.1$\pm $21.8 beats/min. Average conduction times in the PR, the QRS complex and the QTc interval were recorded 103.9$\pm $34.9 msec., 58.3$\pm $23.2 msec. and 302.6$\pm $67.8 msec., in the P and T wave duration recorded 24.8$\pm $6.4 msec. and 51.7$\pm $10.8 msec. respectively. The shape of wave in each leads were observed various types, and any spacial wave type appeared the highest frequency in each lead that ware shown less than 60%, and these frequent rate and average amplitudes as fallow: 1. In P waver the frequent rate and average amplitudes of the positive type showed in leads I, II and aVL that were 54.8% (93.0$\pm $ 33.2 $\mu $V), 50.5% (90.1$\pm $30.5 $\mu $V) and 41.7% (58.5$\pm $ 31.1 $\mu $V). Average amplitude of the negative type showed the frequent rate of 49.8% in lead aVE which was -77.6$\pm$ 25.2 $\mu $V. Biphasic type in leads III and aVF were 46.1% (108.4 $\mu $V, -90.2 $\mu $V.) and 45.7% (137.4 $\mu $V, -105.4 $\mu $V.), and amplitudes between positive and negative of it were significant difference. 2. The highest amplitudes of the QRS complex in all leads were 534.8$\pm $ 232.3 $\mu $V of lead II. The frequent rate and amplitudes of the R wave type in the I, II, III and aVF were 30.2% (277.8 $\pm $131.3 $\mu $V), 45.1% (393.1$\pm $114.2 $\mu $V), 48.5% (349.3$\pm $178.3 $\mu $V) and 54.9% (334.4$\pm $129.7 $\mu $V), and QS ways type in the lead aVL was 49.5% (359.2$\pm $195.5$\mu $V), and RS and QS wave types in the lead a VR were 43.3% (312.4 $\mu $V, -212.7 $\mu $V.) and 41.1% (399.2$\pm $92.2 $\mu $V), respectively. 3. In T wave, the frequent rata and amplitudes of the positive type in the leads I, II, III and aVF were 44.6% (207.9$\pm $ 97.1 $\mu $V), 41.6% (245.1$\pm $92.1 $\mu $V), 46.9% (189.8$\pm $ 82.7 $\mu $V) and 53.0% (195.4197.8 $\mu $V), and the negative in the lead aVR was 41.2% (-230.7$\pm$ 103.1 $\mu $V), respectively. The positive and negative types in the lead aVe appeared with same frequent rate of 43.2%.4. Frontal plane vectors for P, QRS, and T were found to lie at 38.1$\pm $ 21.5, 142.0$\pm $ 57.2, and 117.2$\pm $ 63.9 degrees, respectively. These results in ECG of goats may be served to the limited purposes as to conduction parameters, arrhythmias except abnormal ECG because of waveforms, amplitudes and electrical axis of it were variability.

• 한국지능시스템학회논문지
• /
• 제20권4호
• /
• pp.574-579
• /
• 2010

심전도 신호 잡음 중 기저선 변동 잡음은 신뢰성 있는 심장 질환 진단을 위해 반드시 제거되어야 하는 것으로서, 이를 위해 본 논문에서는 P, T파 및 QRS-complex를 동시에 배제하여 기저선 변동 잡음만을 추정할 수 있는 Morphology-pair를 제안한다. 즉, P, R, T파와 같은 국부 최대값(local maxima) 특성을 가지는 신호 영역과 Q, S파와 같은 국부 최소값(local minima) 특성을 가지는 신호 영역을 배제할 수 있는 각각의 Morphology 연산을 하나의 Morphology-pair로 정의하고, 이를 이용하여 추정된 기저선 변동 잡음 신호와 원 신호와의 차를 통해 기저선 변동 잡음 제거 신호를 도출한다. 제안한 알고리즘의 유효성을 확인하기 위해 실제 심전도 임상 데이터인 MIT/BIH 데이터베이스를 이용한 실험 결과를 살펴봄으로써 기저선 변동 잡음이 효과적으로 제거됨을 입증한다.

• 한국정보통신학회논문지
• /
• 제23권12호
• /
• pp.1542-1550
• /
• 2019

부정맥 분류를 위한 기존 연구들은 분류의 정확성을 높이기 위해 신경회로망(Artificial Neural Network), 퍼지(Fuzzy), 기계학습(Machine Learning) 등을 이용한 방법이 연구되어 왔다. 특히 딥러닝은 신경회로망의 문제인 은닉층 개수의 한계를 해결함으로 인해 오류 역전파 알고리즘을 이용한 부정맥 분류에 가장 많이 사용되고 있다. 딥러닝 모델을 심전도 신호에 적용하기 위해서는 적절한 모델선택과 파라미터를 최적에 가깝게 선택할 필요가 있다. 본 연구에서는 심실 조기 수축 비트 검출을 위한 딥러닝 기반의 최적 파라미터 검출 방법을 제안한다. 이를 위해 먼저 잡음을 제거한 ECG신호에서 R파를 검출하고 QRS와 RR간격 세그먼트를 추출하였다. 이후 딥러닝을 통한 지도학습 방법으로 가중치를 학습시키고 검증데이터로 모델을 평가하였다. 제안된 방법의 타당성 평가를 위해 MIT-BIH 부정맥 데이터베이스를 통해 각 파라미터에 따른 딥러닝 모델로 훈련 및 검증 정확도를 확인하였다. 성능 평가 결과 R파의 평균 검출 성능은 99.77%, PVC는 97.84의 평균 분류율을 나타내었다.

• 대한수의학회지
• /
• 제33권4호
• /
• pp.735-746
• /
• 1993

Electrocardiographic parameters of amplitude and the shape of waves on the unipolar precordial chest leads in the normal Korean native cattles have been measured with a 3 channel electrocardiograph built in a computed and analysis. The study was conducted on 98 heads of mean age of 17.6 months. The wave forms of P, T and QRS complex wave in all leads showed various types. The parameters of the amplitude in the wave types showed the most frequency in each lead that were analyzed as follow : 1. In P wave, amplitudes of positive type showed a frequency of 92.9% and 93.9% in leads $CV_6LU$ and $CV_6LL$ that were $83.8{\pm}31.0{\mu}V$ and $76.0{\pm}30.7{\mu}V$, and negative type showed a frequency of 97.9% in lead V 10 that were $-80.2{\pm}29.4{\mu}V$, respectively. But the plate type in leads $CV_6RU$ and $CV_6RL$ showed frequency of 48% and 58.3%, respectively. 2. Average amplitude of the QRS complex were in a range of $277.0{\pm}154.0{\mu}V$ to $648.2{\pm}146.2{\mu}V$(mean of $418.8{\pm}139.4{\mu}V$) in all leads that were manifested the Low-Voltage QRS complex(below 1 mV in unipolar precordial leads). Average amplitudes of each wave type in the QRS complex were $-250.0{\pm}139.8{\mu}V$ and $-399.2{\pm}226.8{\mu}V$ in the QS group types that showed a frequency of 50.0% and 82.5% in the leads $CV_6LU$ and $CV_6LL$, respectively. And average amplitudes of the R group types showed a frequency of 85.6%, 56.1% and 75.8% in the $CV_6RU$, $CV_6RL$ and $V_{10}$ that were $321.5{\pm}142.1{\mu}V$, $271.6{\pm}139.9{\mu}V$ and $552.4{\pm}132.7{\mu}V$, respectively. 3. In T waves, Amplitudes of the positive type showed a frequency of 60.2%, 46.9% and 83.7% in leads of $CV_6LL$, $CV_6RU$ and $CV_6RL$ that were respectively $184.7{\pm}93.7{\mu}V$ $103.7{\pm}64.43{\mu}V$ and $111.8{\pm}39.3{\mu}V$, the negative type showed a frequency of 62.2% and 93.7% in leads $CV_6LU$ and $V_{10}$ that were $142.2{\pm}82.1{\mu}V$ and $-280.3{\pm}107.2{\mu}V$, respectively. 4. Average amplitude of ST segment were $3.7{\pm}33.1{\mu}V$, $0.9{\pm}23.1{\mu}V$, $10.9{\pm}28.6{\mu}V$, $5.8{\pm}28.3{\mu}V$ and $-34.7{\pm}48.4{\mu}V$ in leads $CV_6LL$, $CV_6RU$, $CV_6RL$, $CV_6LU$ and $V_{10}$ respectively.