• Journal of Sensor Science and Technology
• /
• v.23 no.6
• /
• pp.402-408
• /
• 2014

Heart rate is one of the most important signal to monitor the health condition of the patient or exerciser. Various wearable devices have been developed for the continuous monitoring of ECG signal from human body during exercise. Among these, ECG chest belt has been widely used. However wearing chest belt with ECG sensor is uncomfortable in normal life due to the electrode contact between metal electrodes of ECG sensor and skin of the human body. So we develop the royal healthcare backpack that can measure ECG signal without skin contact by using capacitor-type ECG sensor. The position of the measurement point is critical to collect a clear ECG signal in the capacitive ECG measurement from backpack. Various tests were conducted to find the optimal ECG measurement position which has less noise and could get strong and clear ECG signal during exercise, walking, hiking, mountain climbing and cycling.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.294-294
• /
• 2000

In this paper, we would like to discuss the signal processing and the algorithm for ECG analysis. The ECG gives us information about the condition of the heart muscle, because myocardial abnormality or infarction is inscribed on the ECG during myocardial depolarization and repolarization. Analyzing the ECG signal, we can find heart disease, for example, arrhythmia and myocardial infarction, etc. Particularly, detecting arrhythmia is more important, because serious arrhythmia can take away the life from patients within ten minutes. The wavelet transform decomposes the ECG signal into high and low frequency component using wavelet function. Recomposing high frequency bands including QRS complex, we can detect QRS complex and eliminate the noise from the original ECG signal. To recognize the ECG signal pattern, we adopted the curve-fitting partially and statistical method. The ECG signal is divided into small parts based on QRS complex, and then, each part is approximated to the polynomials. Comparing the approximated ECG pattern with some kinds of heart disease ECG pattern, we can detect and classify the kind of heart disease.

• Proceedings of the KSME Conference
• /
• 2001.06b
• /
• pp.42-47
• /
• 2001

Analyzing the ECG signal, we can find heart disease, for example, arrhythmia and myocardial infarction, etc. Particularly, detecting arrhythmia is more important, because serious arrhythmia can take away the life from patients within ten minutes. In this paper, we would like to introduce the signal processing for ECG analysis and the device made for wireless communication of ECG data. In the signal processing, the wavelet transform decomposes the ECG signal into high and low frequency components using wavelet function. Recomposing the high frequency bands including QRS complex, we can detect QRS complex and eliminate the noise from the original ECG signal. To recognize the ECG signal pattern, we adopted the polynomial approximation partially and statistical method. The ECG signal is divided into small parts based on QRS complex, and then, each part is approximated to the polynomials. Comparing the approximated ECG pattern with the database, we can detect and classify the heart disease. The ECG detection device consists of amplifier, filters, A/D converter and RF module. After amplification and filtering, the ECG signal is fed through the A/D converter to be digitalized. The digital ECG data is transmitted to the personal computer through the RF transceiver module and serial port.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.285-288
• /
• 2008

In ubiquitous computing environment the biological signal ECG (Electrocardiogram signal) is usually recorded with noise components. Adaptive interference (or noise) canceller do adaptive filtering of the noise reference input to maximally match and subtract out noise or interference from the primary (signal plus noise) input thereby adaptively eliminate unwanted interference from the ECG signal. Measured Stress ECG (or exercise ECG signal) signal have three major noisy component like baseline wander noise, motion artifact noise and EMG (Electro-mayo-cardiogram) noise. These noises are not only distorted signal but also root of incorrect diagnosis while ECG data are analyzed. Motion artifact and EMG noises behave like wide band spectrum signals, and they considerably do overlapping with the ECG spectrum. Here the multidimensional adaptive method used for filtering which is more effective to improve signal to noise ratio.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.48-50
• /
• 2006

In this paper, we propose a method to minimize the baseline wandering that make hard to detect R wave in ECG. This method uses a different signal between ECG and ascending slope tracing waves to minimize the baseline wandering. When the slope of ECG signal maintains the value or falls, the ascending slope tracing wave fellows ECG signal directly, and this wave holds that value of ECG signal when the slope begins to rises in a certain time(=hold time). After this hold time, this wave traces ECG signal again. To evaluate this minimizing method for baseline wandering, the experiments are carried out with 5 ECG data in the database of MIT/BIH. R waves in the proposed different signal are detected by using descending slope trace waves and compared with the annotation file. The results show that the proposed method Is sure to minimize the baseline wandering in ECG.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.9
• /
• pp.2361-2376
• /
• 2023

The electrocardiogram (ECG) signal is commonly used to screen and diagnose cardiovascular diseases. In recent years, deep neural networks have been regarded as an effective way for automatic ECG disease diagnosis. The convolutional neural network is widely used for ECG signal extraction because it can obtain different levels of information. However, most previous studies adopt single scale convolution filters to extract ECG signal features, ignoring the complementarity between ECG signal features of different scales. In the paper, we propose a dual-scale network with convolution filters of different sizes for 12-lead ECG classification. Our model can extract and fuse ECG signal features of different scales. In addition, different spatial and time periods of the feature map obtained from the 12-lead ECG may have different contributions to ECG classification. Therefore, we add a spatial-temporal attention to each scale sub-network to emphasize the representative local spatial and temporal features. Our approach is evaluated on PTB-XL dataset and achieves 0.9307, 0.8152, and 89.11 on macro-averaged ROC-AUC score, a maximum F1 score, and mean accuracy, respectively. The experiment results have proven that our approach outperforms the baselines.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.9
• /
• pp.1444-1452
• /
• 1989

This paper describes the implementation of a real time fetal ECG monitoring system in which an adaptive multi-channel noise canceller is realized using the Texas Instruments TMS32020 progrmmmable ditital signal processor. An ECG signal from the electrode placed on the mother's abdomen and three ECGs from those on the chest are applied as the desired signal and the referened inputs, respectively, of the multi-channel filter. The coefficients of the filter are updated using the LMS algorithm such that the output of the multi-channel filter copies the maternal ECG embedded in the abdominal ECG. The enhanced fetal ECG is obtained by subtracting the filter output from the abdominal ECG, and the difference signal is recorded. Both off-line and on-line experimental results are presented to verify the effectiveness of the parameters for the digital signal processing algorithms and the prototype system.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.325-327
• /
• 2004

ECG limb lead II signal widely used to diagnosis heart diseases and it is essential to detect ECG events (onsets, offsets and peaks of the QRS complex P wave and T wave) and extract them from ECG signal for heart diseases diagnoses. However, it is very difficult to develop standardized feature extraction formulas since ECG signals are varying on patients and disease types. In this paper, simple feature extraction method from normal and abnormal types of ECG signals is proposed. As a signal features, heart rate, PR interval, QRS interval, QT interval, interval between S wave and baseline, and T wave types are extracted. To show the validity of proposed method, Right Bundle Branch Block (RBBB), Left Bundle Branch Block (LBBB), Sinus Bradycardia, and Sinus Tachycardia data from MIT-BIH arrhythmia database are used for feature extraction and the extraction results showed higher extraction capability compare to conventional formula based extraction method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.834-837
• /
• 2015

In this paper, we designed the FPGA hardware-based real-time ECG simulator, which generates an analog ECG signal within the range of 0 to 5 volts and described function. The ECG signal generated by the simulator can be applied to laboratory tests, the medical device, and the calibration study in various ways. ECG signals generated by simulator are obtained with conventional 24bit quantization to generate the signal data, and they are sampled and quantized to 1kHz of the 8-bit resolution when used as actual data. The proposed simulator is implemented using xilix Spartan-3 and data are transmitted through an RS-232 between the PC and the FPGA simulator. The transmitted data are stored in the memory and the stored data are printed out with the analog ECG signal through DAC (0808). It can also control the heart rate (HR) via the two buttons level UP-DOWN. We used existing ECG input rating for the evaluation of the designed system and evaluated differential circuit for obtaining QRS waveform and the output signal. We finally could obtained proper the result.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.1 no.1
• /
• pp.7-12
• /
• 2008

This paper studies a method to minimize the baseline wandering that make hard to extract R-wave in ECG. This method uses a difference signal between ECG and ascending slope tracing waves to minimize the baseline wandering. When the slope of ECG signal maintains the value or falls, the ascending slope tracing wave follows ECG signal directly, and this wave holds that value of ECG signal when the slope begins to rises in a certain time(=hold time). After this hold time, this wave traces ECG signal again. This method has been applied to MIT/BIH database to verify its efficacy and validity in practical applications.