Table 1 Pseudocode for SLS trajectories optimization and generation

Input: random path through all ROIs (path₀), labels indicating which ROI corresponds to each path₀ pixel (tags)

Step 1. Find unique ROIs

For ROI in unique(tags):

    Find ROI centroid

Step 2. Find optimal path through ROIs centroids (path_TSP) applying genetic algorithm to solve TSP

Randomly initialize a population of 100 individuals (random paths through all ROIs)

Find minimum path length min_glob and the individual than minimize path length (path_TSP)

For generation in [1:1000]:

    Compute the path length for each individual

    If exist a path shorter then min_glob, then update min_glob and path_TSP

    Randomly split population in groups of 4 individuals

    For each group of 4 individuals:

        Find the individual with shortest path

        Create 4 mutations (original, flip, swap, slide) of the individual for the next generation

Step 3. Compute path_SLS

Initialize empty trajectory path_SLS

For ROI in path_TSP:

    Find path through all ROI’s pixels (path_ROI) using a greedy algorithm

    Append path_ROI to path_TSP

Return: path_TSP