An automated approach for tracking individual nephrons through three-dimensional histological image sets of mouse and rat\nkidneys is presented. In a previous study, the available images were tracked manually through the image sets in order to explore\nrenal microarchitecture.The purpose of the current research is to reduce the time and effort required to manually trace nephrons\nby creating an automated, intelligent system as a standard tool for such datasets. The algorithm is robust enough to isolate closely\npacked nephrons and track their convoluted paths despite a number of nonideal, interfering conditions such as local image\ndistortions, artefacts, and interstitial tissue interference. The system comprises image preprocessing, feature extraction, and a\ncustom graph-based tracking algorithm, which is validated by a rule base and a machine learning algorithm. A study of a selection\nof automatically tracked nephrons, when compared with manual tracking, yields a 95% tracking accuracy for structures in the\ncortex, while those in the medulla have lower accuracy due to narrower diameter and higher density. Limited manual intervention\nis introduced to improve tracking, enabling full nephron paths to be obtained with an average of 17 manual corrections per mouse\nnephron and 58 manual corrections per rat nephron.
Loading....