Detection and Quantification of Fluorescent Cell Clusters in Cryo-Imaging

Table 1

Pseudocode for the determination of model-based parameters and estimation of cell number. Histogram analysis of cluster volume and integrated intensity are used to estimate single cell clusters. Model parameters are determined from the estimated single cells. These parameters are used to estimate the number of cells in a given cluster.

	Preprocessing and determination of model cell parameters
	Load aligned, next image processed cryo-image stack
	Perform morphological reconstruction to remove background
	User interactively determines 𝑇𝑅/𝐺, 𝑇𝐻, and 𝑇𝐿
	Perform thresholding of ratio image
	Perform hysteresis thresholding
	Perform connected component analysis
	Display histogram of cluster volume
	User selects single cell volume range based on largest peak
	Display histogram of integrated intensity
	User selects single cell integrated intensity range based on largest peak
	For all clusters that are within the user defined integrated intensity and volume range
	Fit Gaussian with free parameters (𝑥𝑐,𝑦𝑐,𝑧𝑐), 𝐼𝑡, and 𝜎 to each cluster
	Perform an average of estimated model parameters to determine 𝜎 and 𝐼𝑡
	Estimate number of cells per cluster
	For a given cluster
	Estimate number of cells per cluster
	Test integer numbers of cells within the range
	Perform nonlinear optimization to place the model cells in the 3D positions that best fits the input image
	The number of cells that best fits the inputted image is chosen as the estimated number of cells in the cluster