A segmentation method and classification of diagnosis for thyroid nodules

Abstract

Heterogeneous features of thyroid nodules in ultrasound images is very difficult task whenradiologists and physicians manually draw a complete shape of nodule, size and shape, image or distinguish what type of nodule is exist. Segmentation and classification is important methods for medical image processing. Ultrasound imaging is the best way to prediction of which type of thyroid is there. In this paper, uses the groups Benign (non-cancerous) and Malignant (cancerous) Thyroid Nodules images were used. The texture feature method like GLCM are very useful for classifying texture of images and these features are used to train the classifiers such as SVM, KNN and Bayesian. The experimental result shows the performance of the classifiers and shows the best predictive value and positively identify the percentage of the non-cancerous or cancerous people and shows the best performance accuracy using the SVM classifier as compare to the KNN and Bayesian classifier.

Improving Ultrasound Image Classification with Local Texture Quantisation

“…Work [2,3] collected large-scaled datasets and trained deep learning models on them; however, these datasets are not publicly available. Because of these limitations and challenges, many recent works [4,5,6] used shallow models like SVM, KNN, and logistic regression rather than deep learning based models. Some other work [7] manually pre-processed the images to reduce noises before a neural network model.…”

Multi‐objectives optimisation of features selection for the classification of thyroid nodules in ultrasound images

“…Singh and Jindal [3] have developed a method for the classification of thyroid nodules based on texture analysis. The grey‐level co‐occurrence matrix (GLCM) was used for the extraction of contrast, correlation, energy, solidity, entropy, variance and homogeneity.…”

“…Several medical imaging tools can be used to assess thyroid nodules. Ultrasound (US) imaging is highly recommended to assess this disease at an early stage, because of the significant advantages it presents [2, 3] such as the low cost, the absence of any ionising radiation, it is non‐invasive and makes it easier to distinguish between benign and malignant nodules etc. Certain characteristics [4] are generally used by radiologists to differentiate these nodules.…”

“…CAD systems dedicated to the classification of thyroid nodules, generally ignore the problem of redundant and irrelevant characteristics. In addition and very often, during the evaluation phase of many of these systems, a limited set of ultrasound image data is used [3, 15, 23]. The aim of this work is to propose a novel approach for the classification of the thyroid nodules in the US using the statistical features.…”

Focal and diffused liver disease classification from ultrasound images based on isocontour segmentation

“…Since most of the liver ultrasound images consist of irregular and diffused regions, texture‐based methods are found to be more suitable to recognise the patterns spread out in the image, rather than the Euclidean shape‐based techniques. Of the various texture‐based approaches available in the literature [28], the Grey level co‐occurrence Matrix (GLCM) is found to be more suitable for extracting texture features from ultrasound images [29–32]. Recently, Fractal‐based techniques have been analysed for texture feature extraction on lung CT datasets [33, 34].…”