Study of T-wave End Detection with Geometric Method
	    		
		   		
		   			
		   		
	    	
    	 
    	10.3969/j.issn.1005-202X.2010.02.013
   		
        
        	
        		- VernacularTitle:基于几何算法的T波终点检测研究
 
        	
        	
        	
        		- Author:
	        		
		        		
		        		
			        		Juan WANG
			        		
			        		;
		        		
		        		
		        		
			        		Zhongchao HUANG
			        		
			        		;
		        		
		        		
		        		
			        		Zhengchun LIU
			        		
			        		
		        		
		        		
		        		
		        		
		        		
			        		
			        		
		        		
	        		
        		 
        	
        	
        	
        		- Publication Type:Journal Article
 
        	
        	
        		- Keywords:
        			
	        			
	        				
	        				
			        		
				        		T-wave end;
			        		
			        		
			        		
				        		wavelet transforming;
			        		
			        		
			        		
				        		geometrical method;
			        		
			        		
			        		
				        		detection interval
			        		
			        		
	        			
        			
        		
 
        	
            
            
            	- From:
	            		
	            			Chinese Journal of Medical Physics
	            		
	            		 2010;27(2):1751-1754
	            	
            	
 
            
            
            	- CountryChina
 
            
            
            	- Language:Chinese
 
            
            
            	- 
		        	Abstract:
			       	
			       		
				        
				        	Purpose: To realize the detection of T-wave end with an efficient and accurate method. Methods: R-wave was lo-cated using wavelet transforming, and then the T-wave detection interval was identified accttrately by avoiding the influence of heart rate and P-wave. In addition, T-wave peaks were determined referring to the modulus maxima in the interval. Based on these, the T-wave end was located with a simple and practical geometrical method, which was applied for the evaluation of QT database. Results: The positioning capability of the above-mentioned method reached to (-0.35225±18.5869) ms which was comparable or even better than those manual annotations from QT database by cardiologists. Conclusions: The experimental results showed that our algorithm for T-wave end detection is robust to noise, baseline drift and waveform morphological varia-tions, and easy to calculate and to realize.