基于k-means的纸质心电图数字化信息提取研究

针对临床诊断中纸质心电图容易遭到不同程度的损坏,且不便于保存的问题,探讨了纸质心电图的数字化问题。采用均值法降噪,在Sobel算法中加入了45°和135°2个方向的模板,利用k-means方法进行了改进,从而得到更加精准的心电图的数字化提取图形。实际验证发现,心电图的数字化提取精度得到了有效提升,有望在医疗实际中推广应用。

Research of digital information extraction from paper ECG based on the k-means algorithm

Aiming at the problems of paper ECG in the clinical diagnosis, such as vulnerable to suffer different degrees of damage and inconvenient to save, the digitization of paper ECG is discussed. Averaging method is adopted to reduce noise, templates in two directions at 45° and 135°are added into Sobel algorithm, and the algorithm is improved by k-means method, then more accurate extracted ECG digital graphics are obtained. Actual verification shows that the digital extraction accuracy of ECG is promoted effectively, and the method is expected to be popularized and applied in medical practice.