The thyroid is one of the largest endocrine glands in the human body, which is involved in several body mechanisms like controlling protein synthesis and the body's sensitivity to other hormones and use of energy sources. Hence, it is of prime importance to track the shape and size of thyroid over time in order to evaluate its state. Thyroid segmentation and volume computation are important tools that can be used for thyroid state tracking assessment. Most of the proposed approaches are not automatic and require long time to correctly segment the thyroid. In this work, we compare three different nonautomatic segmentation algorithms (i.e., active contours without edges, graph cut, and pixel-based classifier) in freehand three-dimensional ultrasound imaging in terms of accuracy, robustness, ease of use, level of human interaction required, and computation time. We figured out that these methods lack automation and machine intelligence and are not highly accurate. Hence, we implemented two machine learning approaches (i.e., random forest and convolutional neural network) to improve the accuracy of segmentation as well as provide automation. This comparative study intends to discuss and analyse the advantages and disadvantages of different algorithms. In the last step, the volume of the thyroid is computed using the segmentation results, and the performance analysis of all the algorithms is carried out by comparing the segmentation results with the ground truth.

No abstract

Ultrasound (US) is a widely used as a low-cost alternative to computed tomography (CT) or magnetic resonance (MRI) and primarily for preliminary imaging. Since speckle intensity in US images is inherently stochastic, readers are often challenged in their ability to identify the pathological regions in a volume of a large number of images. This paper introduces a generalized approach for volumetric segmentation of structures in US images and volumes. We employ an iterative random walks (IRW) solver, random forest (RF) learning model, and a gradient vector flow (GVF) based inter-frame belief propagation technique for achieving cross-frame volumetric segmentation. At the start, a weak estimate the tissue structure is obtained using estimates of parameters of a statistical mechanics model of ultrasound tissue interaction. Ensemble learning of these parameters further using a random forest is used to initialize the segmentation pipeline. IRW is used for correcting the contour in various steps of the algorithm. Subsequently, a GVF based inter-frame belief propagation is applied to adjacent frames based on initialization of contour using information in the current frame to segment the complete volume by frame-wise processing. We have experimentally evaluated our approach using two different datasets. Intravascular Ultrasound (IVUS) segmentation was evaluated using 10 pullbacks acquired at 20 MHz and thyroid ultrasound segmentation is evaluated on 16 volumes acquired at 11 - 16 MHz. Our approach obtains a Jaccard score of 0.937 ± 0.022 for IVUS segmentation and 0.908 ± 0.028 for thyroid segmentation while processing each frame in 1.15 ± 0.05 s for IVUS and in 1.23 ± 0.27 s for thyroid segmentation without the need of any computing accelerators like GPUs.

Texture analysis is an important topic in Ultrasound (US) image analysis for structure segmentation and tissue classification. In this work a novel approach for US image texture feature extraction is presented. It is mainly based on parametrical modelling of a signal version of the US image in order to process it as data resulting from a dynamical process. Because of the predictive characteristics of such a model representation, good estimations of texture features can be obtained with less data than generally used methods require, allowing higher robustness to low Signal-to-Noise ratio and a more localized US image analysis. The usability of the proposed approach was demonstrated by extracting texture features for segmenting the thyroid in US images. The obtained results showed that features corresponding to energy ratios between different modelled texture frequency bands allowed to clearly distinguish between thyroid and non-thyroid texture. A simple k-means clustering algorithm has been used for separating US image patches as belonging to thyroid or not. Segmentation of thyroid was performed in two different datasets obtaining Dice coefficients over 85%.

The thyroid is one of the largest endocrine glands in the human body, which is involved in several body mechanisms like controlling protein synthesis, use of energy sources, and controlling the body's sensitivity to other hormones. Thyroid segmentation and volume reconstruction are hence essential to diagnose thyroid related diseases as most of these diseases involve a change in the shape and size of the thyroid over time. Classification of thyroid texture is the first step toward the segmentation of the thyroid. The classification of texture in thyroid Ultrasound (US) images is not an easy task as it suffers from low image contrast, presence of speckle noise, and non-homogeneous texture distribution inside the thyroid region. Hence, a robust algorithmic approach is required to accurately classify thyroid texture. In this paper, we propose three machine learning based approaches: Support Vector Machine; Artificial Neural Network; and Random Forest Classifier to classify thyroid texture. The computation of features for training these classifiers is based on a novel approach recently proposed by our team, where autoregressive modeling was applied on a signal version of the 2D thyroid US images to compute 30 spectral energy-based features for classifying the thyroid and non-thyroid textures. Our approach differs from the methods proposed in the literature as they use image-based features to characterize thyroid tissues. We obtained an accuracy of around 90% with all the three methods. INDEX TERMSMedical imaging, support vector machine, artificial neural network, random forest classifier, texture classification, thyroid ultrasound. During the course of his B.Sc., he worked as a Research Assistant with Fraunhofer Mevis, Bremen, Germany, and recently he was a Visiting Researcher with General Electric Healthcare, Milwaukee, USA. His research interest includes medical image processing, computer vision, and machine learning.

In this paper, we propose a method to segment the thyroid from a set of 2D ultrasound images. We extended an active contour model in 2D to generate a 3D segmented thyroid volume. First, a preprocessing step is carried out to suppress the noise present in US data. Second, an active contour is used to segment the thyroid in each of the 2D images. Finally, all the segmented thyroid images are passed to a 3D reconstruction algorithm to obtain a 3D model of the thyroid. We obtained an average segmentation accuracy of 86.7% in six datasets with a total of 703 images.

No abstract

No abstract