Average Radial Length Crossing Code for Spheroid Invasion Analysis

The script AverageRadialLengthCrossings.m was developed for shape factor analysis of spheroids and organoids as featured in the manuscript “Shape Factor Analysis as a Quantitative Framework for Assessing Spheroid and Organoid Morphology and Invasiveness”. AverageRadialLengthCrossings.m analyses ImageJ/FIJI regions of interest and computes the radial length from the centroid to each perimeter point of the region of interest. Then the code plots perimeter point number vs. radial length overlayed with the average radial length to compute the number of Average Radial Length Crossings (ARLC).

Breast cancer is diagnosed in 1 of every 8 women over the course of their lifetime with most detection coming from imaging-based screenings. In the clinic, radiologists determine if a breast tumor is invasive through a visual assessment of mammogram/ultrasound images. Over the past decades there have been efforts to develop computer added detection methods to assist radiologists in identifying invasive tumors.   Kilday et. al. proposed radial length crossing metrics for mammogram analysis. As complex in vitro models are increasingly used for drug development and pharmaceutical research, bench scientists need a quantitative metric to assess invasion in in vitro systems without undergoing the same level of image interpretation training as radiologist.  Inspired by computer assistive diagnostic technologies, we developed this MATLAB code to take a segmented spheroid/organoid region of interest and quantify invasive phenotype via computing the number of average radial length crossings.