1. How to download the data to get the MIT-BIH
Download all 48 groups of MIT-BIH ECG data from the following official link page:
Http://www.physionet.org/physiobank/database/mitdb/
The following link is a detailed description of the MIT-BIH database and user guide, good English students may wish to carefully read:
Http://www.physionet.org/physiobank/database/html/mitdbdir/mitdbdir.htm
Ii. MIT-BIH Data Format
To save the file length and storage space, MIT uses a custom format, so there is no common reading method.
An ECG record consists of three parts:
(1) the header file [. Hea] is stored in ASCII code characters.
(2) Data Files [. dat] are stored in binary format. Each three bytes store two numbers, one of which is 12bit.
(3) Comment on the file [. art], stored in binary format, the definition is complex, but as long as you know how to read the second (2) into MATLAB, you can introduce how to read the second (3.
3. Reading MIT-BIH data using tools on the official website
You can use this link to browse various physiological databases of MIT:
Http://www.physionet.org/cgi-bin/chart
In addition, the tool provided in the upper-right corner of the webpage converts signals to TXT files. The specific figure is as follows:
(1) After Entering the http://www.physionet.org/cgi-bin/chart page, select the desired database from the drop-down menu at the top of the page, such as MIT-BIH arrhythmia database (mitdb )", click the "continue" button next to the drop-down menu to go to the next page;
(2) Select the expected ECG data from the drop-down menu corresponding to the record, for example, "100". To view the signal waveform, click the "show Chart" button below, you can see a 10-second ECG. You can click the arrow on the left or right of the page as needed to view the ECG waveform of the first 10 seconds or the next 10 seconds;
(3) to obtain data in document format, find the "Convert signals to text" link in the "record" line at the rightmost of the page, and click the link to go to the next page; then, click the show samples button on the page to obtain the time information of the record and the specific data of the two groups of signals from 0 to 60 seconds:
(4) Of course, the website provides us with a convenient data email sending function. You only need to enter your email address after the "E-mail samples to" button, and then click this button, you can send the required ECG record data to your mailbox.