Electronic device and method for selecting feature of electrocardiogram
An electronic device and a method for selecting a feature of an electrocardiogram (ECG) are provided. The method includes: obtaining the ECG; performing a first pre-processing on the ECG to generate a first ECG; marking multiple extreme points corresponding to at least one type of wave on the first ECG; calculating a first feature value corresponding to a first feature according to the multiple extreme points of the at least one type of wave, generating a first performance index corresponding to a machine learning model according to the first feature value, and determining whether to select the first feature according to the first performance index; and outputting the first feature in response to selecting the first feature.
1. An electronic device for selecting a feature of an electrocardiogram, comprising:
a transceiver;
a storage medium, storing a plurality of modules; and
a processor, coupled to the storage medium and the transceiver, and accessing and executing the plurality of modules, wherein the plurality of modules comprise:
a data collection module, obtaining the electrocardiogram through the transceiver;
a first pre-processing module, performing a first pre-processing on the electrocardiogram to generate a first electrocardiogram;
a second pre-processing module, performing a second pre-processing on the first electrocardiogram to generate a second electrocardiogram;
a marking module, marking a plurality of extreme points corresponding to at least one type of wave comprising an R wave on the first electrocardiogram by: dividing the second electrocardiogram into a plurality of portions according to a window function; marking a data point with a maximum value in a first portion in the plurality of portions as a first reference point in response to the first portion comprising at least one data point greater than zero, wherein the first reference point corresponds to a first reference time point; determining a first period of the first electrocardiogram according to the first reference time point; and marking a second data point with a maximum value in the first period as a first R wave extreme point, wherein the first reference time point is located at a center of the first period, wherein the plurality of extreme points comprise the first R wave extreme point;
a feature establishing module, calculating a first feature value corresponding to a first feature according to the plurality of extreme points of the at least one type of wave, and generating a first performance index corresponding to a machine learning model according to the first feature value;
a feature selection module, determining whether to select the first feature according to the first performance index; and
an output module, outputting the first feature through the transceiver in response to selecting the first feature.
2. The electronic device according to claim 1 , wherein the at least one type of wave further comprises a Q wave, wherein the first R wave extreme point corresponds to a first time point, wherein
the marking module determines a second period of the first electrocardiogram according to the first time point, and marks a third data point with a minimum value in the second period as a first Q wave extreme point, wherein the first time point is a latest time point of the second period, wherein the plurality of extreme points comprise the first Q wave extreme point.
3. The electronic device according to claim 2 , wherein
the marking module marks a fourth data point with a minimum value in a third period earlier than the second period as the first Q wave extreme point in response to a slope of the third data point being positive.
4. The electronic device according to claim 2 , wherein the first Q wave extreme point corresponds to a second time point, wherein
the marking module marks a plurality of R wave extreme points comprising the first R wave extreme point on the first electrocardiogram, calculates an average RR interval according to the plurality of R wave extreme points, determines a weight according to a number of the plurality of R wave extreme points, and determines a second window function according to the average RR interval and the weight, wherein
the marking module determines a fourth period earlier than the second time point according to the second window function, and marks a fifth data point with a maximum value in the fourth period as a first P wave extreme point, wherein the second time point corresponds to a second latest time point of the fourth period, wherein the plurality of extreme points comprise the first P wave extreme point.
5. The electronic device according to claim 4 , wherein
the marking module marks a sixth data point with a maximum value in a fifth period earlier than the fourth period as the first P wave extreme point in response to a slope of the fifth data point being negative.
6. The electronic device according to claim 1 , wherein the first pre-processing comprises:
performing baseline wandering removal, noise removal, and standardization on the electrocardiogram to generate the first electrocardiogram.
7. The electronic device according to claim 1 , wherein the second pre-processing comprises:
dividing the first electrocardiogram into three portions, wherein the three portions comprise a first period electrocardiogram, a second period electrocardiogram later than the first period electrocardiogram, and a third period electrocardiogram later than the second period electrocardiogram, wherein lengths of the first period electrocardiogram and the third period electrocardiogram are the same;
calculating a first standard deviation corresponding to the first period electrocardiogram and a second standard deviation corresponding to the second period electrocardiogram; and
deleting the first period electrocardiogram from the first electrocardiogram in response to a difference value between the first standard deviation and the second standard deviation being greater than a first threshold.
8. The electronic device according to claim 7 , wherein the second pre-processing further comprises:
performing wavelet transform on the first electrocardiogram to generate a transformed electrocardiogram;
standardizing the transformed electrocardiogram into a standard score; and
setting a plurality of data points less than a second threshold in the standard score to zero to generate the second electrocardiogram.
9. The electronic device according to claim 1 , wherein
the data collection module receives a plurality of electrocardiograms through the transceiver, wherein the plurality of electrocardiograms comprise a first lead electrocardiogram and a second lead electrocardiogram, wherein
the data collection module standardizes the first lead electrocardiogram into a first standard score function, and selects at least one first standard score from a plurality of first standard scores of the first standard score function in response to at least one first absolute value of the at least one first standard score in the first standard score function being greater than a first absolute value of a first standard score in the first standard score function to calculate a first total, wherein
the data collection module standardizes the second lead electrocardiogram into a second standard score function, and selects at least one second standard score from a plurality of second standard scores of the second standard score function in response to at least one second absolute value of the at least one second standard score in the second standard score function being greater than a second absolute value of a second standard score in the second standard score function to calculate a second total, wherein
the data collection module selects the electrocardiogram corresponding to a third lead from the plurality of electrocardiograms according to signs of the first total and the second total.
10. The electronic device according to claim 9 , wherein the first lead electrocardiogram corresponds to a lead I, wherein the second lead electrocardiogram corresponds to a lead aVF, wherein the third lead corresponds to one of a lead II, a lead aVL, the lead aVF, and a lead aVR.
11. The electronic device according to claim 1 , wherein
the feature selection module selects the first feature in response to the first performance index being greater than a performance threshold.
12. The electronic device according to claim 1 , wherein
the feature establishing module calculates a second feature value corresponding to a second feature according to the plurality of extreme points of the at least one type of wave, and generates a second performance index corresponding to the machine learning model according to the second feature value, wherein
the feature selection module selects the first feature from the first feature and the second feature in response to the first performance index being greater than the second performance index.