A Methodology and Metric for Quantitative Analysis and Parameter Optimization of Unsupervised, Multi-region Image Segmentation

While image segmentation makes up a vital step in the process of such tasks in the medical domain as tissue classification, content-based image retrieval, and computer-aided diagnosis, it remains an area of much debate regarding how one interprets the results of machine segmented regions. Many segmentation methods are still evaluated using a subjective human opinion of quality with a lack of quantitative analysis. Ideally, segmentation would be performed on an image with as little aid from a human user as possible, so solid quantitative analysis of results and optimization of user-defined parameters are a must. This paper proposes the use of a methodology based on eight individual performance measures. It then introduces a metric based on a statistical analysis of the overlap between machine segmented and corresponding ground truth images to evaluate and optimize algorithm parameters, and compare inter-algorithm performance for unsupervised segmentation algorithms.