Hintermüller Christoph, Seger Michael, Pfeifer Bernhard, Fischer Gerald, Modre Robert, Tilg Bernhard
Institute of Biomedical Engineering, University of Health Sciences, Medical Informatics and Technology, Eduard Wallnöfer Zentrum I, 6060 Hall i. Tyrol, Austria.
IEEE Trans Biomed Eng. 2006 Oct;53(10):2055-66. doi: 10.1109/TBME.2006.881797.
Methods for noninvasive imaging of electric function of the heart might become clinical standard procedure the next years. Thus, the overall procedure has to meet clinical requirements as an easy and fast application. In this paper, we propose a new electrode array which improves the resolution of methods for activation time imaging considering clinical constraints such as easy to apply and compatibility with routine leads. For identifying the body-surface regions where the body surface potential (BSP) is most sensitive to changes in transmembrane potential (TMP), a virtual array method was used to compute local linear dependency (LLD) maps. The virtual array method computes a measure for the LLD in every point on the body surface. The most suitable number and position of the electrodes within the sensitive body surface regions was selected by constructing effort gain (EG) plots. Such a plot depicts the relative attainable rank of the leadfield matrix in relation to the increase in number of electrodes required to build the electrode array. The attainable rank itself was computed by a detector criterion. Such a criterion estimates the maximum number of source space eigenvectors not covered by noise when being mapped to the electrode space by the leadfield matrix and recorded by a detector. From the sensitivity maps, we found that the BSP is most sensitive to changes in TMP on the upper left frontal and dorsal body surface. These sensitive regions are covered best by an electrode array consisting of two L-shaped parts of approximately 30 cm x 30 cm and approximately 20 cm x 20 cm. The EG analysis revealed that the array meeting clinical requirements best and improving the resolution of activation time imaging consists of 125 electrodes with a regular horizontal and vertical spacing of 2-3 cm.
心脏电功能的无创成像方法可能在未来几年成为临床标准程序。因此,整个程序必须满足临床要求,即应用简便快捷。在本文中,我们提出了一种新的电极阵列,考虑到易于应用和与常规导联兼容等临床限制,该阵列提高了激活时间成像方法的分辨率。为了识别体表电位(BSP)对跨膜电位(TMP)变化最敏感的体表区域,使用虚拟阵列方法计算局部线性相关性(LLD)图。虚拟阵列方法计算体表每个点的LLD度量。通过构建功效增益(EG)图,选择敏感体表区域内最合适的电极数量和位置。这样的图描绘了导联场矩阵的相对可达到秩与构建电极阵列所需电极数量增加之间的关系。可达到秩本身是通过一个检测准则计算出来的。这样的准则估计当通过导联场矩阵映射到电极空间并由检测器记录时,未被噪声覆盖的源空间特征向量的最大数量。从灵敏度图中,我们发现BSP在左上额和背部体表对TMP变化最敏感。这些敏感区域最好由一个由两个大致为30 cm×30 cm和20 cm×20 cm的L形部分组成的电极阵列覆盖。EG分析表明,最符合临床要求并提高激活时间成像分辨率的阵列由125个电极组成,电极的水平和垂直间距为2 - 3 cm且规则排列。