The movement of the cells and their interactions are geometrically modelled using a directed graph, where the nodes (V) represent the detections and the edges (E) connect spatiotemporally close detections. Each node contains features (orange squares) such as the cell’s centroid and some relevant descriptors (for example, the cell’s morphological and intensity attributes). The edges contain features (blue squares) too, in this case encoding the Euclidean distance between the centroids of the cells. In this example, the node of interest, labelled with the subindex i , is connected to neighbouring nodes in the future, labelled with the subindex j within a distance-based likelihood radius (the edge between nodes with feature vectors vi and vj = 4 is dumped). Meaningful biological events (for example, cell divisions) are naturally encoded in the graph.
