Thyroid classification and segmentation in ultrasound images using machine learning algorithms

The clinical reports usually offer morphometric data in terms of change relative to a prior study. Therefore, to provide the information about an object clinically in terms of its size and shape, image segmentation and classification are important tools in medical image processing. Ultrasound is a versatile imaging technique that can reveal the internal structure of organs, often with astounding clarity. Ultrasound is unique in its ability to image patient anatomy and physiology in real time, providing an important, rapid and non-invasive means of evaluation. Ultrasound continues to make significant contributions to patient care by reassuring patients and enhancing their quality of life by helping physicians understand their anatomy in ways not possible with other techniques. US imaging is thus one of the most commonly used auxiliary tools in clinical diagnosis. In this paper, an automatic system is developed that classifies the thyroid images and segments the thyroid gland using machine learning algorithms. The classifiers such as SVM, ELM are used. The features such as mean, variance, Coefficient of Local Variation Feature, Histogram Feature, NMSID Feature, and Homogeneity are extracted and these features are used to train the classifiers such as ELM and SVM. The results are compared with the ground truth images obtained from the radiologist and the performance measure such as accuracy is evaluated. It is observed that the segmentation using ELM is better than SVM classifier.