<![CDATA[The pumping of blood throughout the body is carried out by the most important organ in the body, namely the heart. Because the heart is the most important organ, the higher the risk of a disorder, one of these disorders is Sleep Apnea. Sleep Apnea is a respiratory disorder that causes breathing to stop momentarily for several times during sleep. Sleep apnea detection can be done in various ways, one of which is by means of a device called an Electrocardiogram (ECG). The way this tool works is by recording the signal issued by the heart when it beats. Currently, to detect Sleep Apnea requires a large amount of money and can only be done in a hospital. For this reason, a sleep apnea detection study was created which allows users to not have to pay expensive fees and can be done anywhere. The features for the detection of Sleep Apnea in this study use the value of the QRS Interval and the Duration P of the ECG signal generated. With the AD8232 sensor and 3 electrodes attached where 2 are attached to the chest and 1 electrode on the abdomen, the ECG signal will be detected. Signal processing and classification are carried out when signal data is obtained from the sensor using the Support Vector Machine (SVM) classification on the Arduino Uno microcontroller. A total of 48 training data and 24 test data were used for determining and testing the accuracy of the SVM. 20 normal data and 28 Sleep Apnea data were used as training data and 10 normal data and 14 Sleep Apnea data were used as test data. The “Normal†class or “Sleep Apnea†class is obtained and displayed when the SVM classification results have been assigned. 83.3% of accuracy is obtained from the SVM classification trial which has an average computational training time of 9.78 seconds and an average computation testing of 0.7 seconds.]]>
Copyrights © 2021
