Application of Karhunen-Loeve transform in nuclear cardiology: Spatio-temporal smoothing and quantitative image analysis

Advances in computer technology and software have created new opportunities for nuclear cardiology. Image analysis is becoming more and more rapid and easier to perform, and many different methods of imaging are available now. Given this situation the diagnostic value of any proposed method is the main criteria for its usefulness. Previous studies showed the superiority of Karhunen-Loeve transform (KLT) over fist harmonic Fourier analysis - currently the method most often used for parametric imaging in RNV. In this study we revised our previous results concerning the accuracy of spatio-temporal smoothing of image sequences by using KLT and assessed the potential of exercise RNV combined with KLT for quantitative analysis of left ventricular contraction.