Generation of digital time database from paper ECG records and Fourier transform-based analysis for disease identification.

ECG signals recorded on paper are transferred to the digital time database with the help of an automated data extraction system developed here. A flatbed scanner is used to form an image database of each 12-lead ECG signal. Those images are then fed into a Pentium PC having a system to extract pixel-to-pixel co-ordinate information to form a raw database with the help of some image processing techniques. These raw data are then ported to the regeneration domain of the system to check the captured pattern with the original wave shape. The sampling period of each ECG signal is computed after detection of QRS complex. Finally, discrete Fourier transform of the generated database is performed to observe the frequency response properties of every ECG signal. Some interesting amplitude properties of monopolar chest lead V4 and V6 are noticed which are stated.