Computer algorithms for automated detection and analysis of local Ca2+ releases in spontaneously beating cardiac pacemaker cells

Local Ca2+ Releases (LCRs) are crucial events involved in cardiac pacemaker cell function. However, specific algorithms for automatic LCR detection and analysis have not been developed in live, spontaneously beating pacemaker cells. In the present study we measured LCRs using a high-speed 2D-camera in spontaneously contracting sinoatrial (SA) node cells isolated from rabbit and guinea pig and developed a new algorithm capable of detecting and analyzing the LCRs spatially in two-dimensions, and in time. Our algorithm tracks points along the midline of the contracting cell. It uses these points as a coordinate system for affine transform, producing a transformed image series where the cell does not contract. Action potential-induced Ca2+ transients and LCRs were thereafter isolated from recording noise by applying a series of spatial filters. The LCR birth and death events were detected by a differential (frame-to-frame) sensitivity algorithm applied to each pixel (cell location). An LCR was detected when its signal changes sufficiently quickly within a sufficiently large area. The LCR is considered to have died when its amplitude decays substantially, or when it merges into the rising whole cell Ca2+ transient. Ultimately, our algorithm provides major LCR parameters such as period, signal mass, duration, and propagation path area. As the LCRs propagate within live cells, the algorithm identifies splitting and merging behaviors, indicating the importance of locally propagating Ca2+-induced-Ca2+-release for the fate of LCRs and for generating a powerful ensemble Ca2+ signal. Thus, our new computer algorithms eliminate motion artifacts and detect 2D local spatiotemporal events from recording noise and global signals. While the algorithms were developed to detect LCRs in sinoatrial nodal cells, they have the potential to be used in other applications in biophysics and cell physiology, for example, to detect Ca2+ wavelets (abortive waves), sparks and embers in muscle cells and Ca2+ puffs and syntillas in neurons.局部钙释放（Local Ca2+ Releases, LCRs）是参与心脏起搏细胞功能的核心事件。然而，针对可自发搏动的活起搏细胞的自动LCR检测与分析专用算法尚未被开发。在本研究中，我们采用高速二维相机对从家兔和豚鼠分离的、可自发收缩的窦房结（sinoatrial, SA）细胞中的LCRs进行了检测，并开发了一种可在时空二维层面检测与分析LCRs的全新算法。本算法追踪收缩细胞中线的各点位，并将这些点位作为仿射变换的坐标系，生成细胞不再收缩的变换图像序列。随后，通过一系列空间滤波器将动作电位诱发的钙瞬变与LCRs从记录噪声中分离出来。通过对每个像素（细胞位置）应用差分（逐帧）灵敏度算法，可检测LCR的产生与消亡事件：当某信号在足够大的区域内变化足够迅速时，即可判定为LCR；当该信号的振幅显著衰减，或其融入整体细胞钙瞬变的上升相时，则认为该LCR已消亡。最终，本算法可输出LCR的核心参数，包括周期、信号总量、持续时长以及传播路径面积。由于LCRs在活细胞内传播，本算法可识别其分裂与融合行为，这表明局部传播的钙诱发钙释放（Ca2+-induced-Ca2+-release, CICR）对于LCR的命运以及产生强大的整体钙信号至关重要。因此，我们的全新计算机算法可消除运动伪影，并从记录噪声与全局信号中检测出二维局部时空事件。尽管本算法是为检测窦房结细胞中的LCRs而开发的，但其有望应用于生物物理学与细胞生理学的其他场景，例如检测肌细胞中的钙小波（abortive waves）、钙火花与钙余烬，以及神经元中的钙脉冲与syntillas。