| Summary: | Cardiac electrical signal or known as electrocardiography (ECG) can be
used to indicate the abnormality. By observing the characteristics of the signal,
cardiac abnormalities can be identified. A microcontroller used as a device to
monitor heart failure in patients is easy to carry and has an ability to quickly and
accurately identify cardiac abnormalities. A set of testing model is needed to
evaluate cardiac abnormalities identification system based microcontroller with
an authentic signal source. A testing model is consisting of ECG signal database,
analog signal generator, and microcontroller based system which is containing
algorithms to detect cardiac abnormalities.
MIT-BIH database provides authentic ECG signal data that can be used
as a source to test the system with varied type of disorders and duration of
observation. MIT-BIH ECG signals are converted to analog signals using 11-bit
DAC with 360 Hz frequency conversion. Microcontroller converts the analog
signals from the output of the generator using an internal 10-bit ADC with a
sampling frequency of 200 Hz. Cardiac abnormalities are then analysed based on
data sampling. Abnormal heart rhythms are identified using R peak parameter. By
measuring the interval between R peaks, the number of beats per minute (bpm)
and the interval variation between R peaks can measured to determine abnormal
heart rhythms.
Results show that DAC output obtains error range from 6.72 milivolt to
14.58 milivolt, whereas ADC output obtains error range from 1 bit to 2 bit.
Statistically, test results show significance values from ideal values are greater
than α = 0,05 meaning that there is no significant difference between measured R-
R intervals with the original R-R intervals by 95% confidence level. The test
method successfully detects multiple type of heart rhythms with category: normal,
bradycardia, tachycardia, and irregular.
|
|---|