Image texture analysis of sonograms in chronic inflammations of thyroid gland

Smutek, Daniel; Šára, Radim; Sucharda, P; Tjahjadi, Tardi; Švec, Martin

doi:10.1016/s0301-5629(03)01049-4

Cited by 52 publications

(34 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The simplest explanation is that the LT class is represented by several subclasses in our data and has thus multi-modal probability distribution function. We observed this property in our earlier experiments as well [18]. We can see that the leave-one-out error L increases initially and then drops back to the 8.1% level.…”

Section: Full Search In Low-dimensional Spacementioning

confidence: 53%

“…Further step of our research focused on the selection of a subset of co-occurrence matrix features suitable for classification [17]. In [18] we tried to use a large set of texture features based on co-occurrence matrices combined with features proposed by Muzzolini et al [14]. As there was no way to systematically explore all possible features to find the best-suited, we turned our attention towards methods that do not require classical features.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Systematic Construction of Texture Features for Hashimoto’s Lymphocytic Thyroiditis Recognition from Sonographic Images

Šára

Smutek

Sucharda

et al. 2001

Artificial Intelligence in Medicine

Self Cite

View full text Add to dashboard Cite

Abstract. The success of discrimination between normal and inflamed parenchyma of thyroid gland by means of automatic texture analysis is largely determined by selecting descriptive yet simple and independent sonographic image features. We replace the standard non-systematic process of feature selection by systematic feature construction based on the search for the separation distances among a clique of n pixels that minimise conditional entropy of class label given all data. The procedure is fairly general and does not require any assumptions about the form of the class probability density function. We show that a network of weak Bayes classifiers using 4-cliques as features and combined by majority vote achieves diagnosis recognition accuracy of 92%, as evaluated on a set of 741 B-mode sonographic images from 39 subjects. The results suggest the possibility to use this method in clinical diagnostic process.

show abstract

Section: Full Search In Low-dimensional Spacementioning

confidence: 53%

Section: Introductionmentioning

confidence: 99%

Systematic Construction of Texture Features for Hashimoto’s Lymphocytic Thyroiditis Recognition from Sonographic Images

Šára

Smutek

Sucharda

et al. 2001

Artificial Intelligence in Medicine

Self Cite

View full text Add to dashboard Cite

show abstract

“…Effect of gain settings have been reported previously on texture properties of the image in various studies [16,33]. However, there are no studies that determine the differences in texture parameter due to gain settings.…”

Section: Discussionmentioning

confidence: 96%

“…Analysis of images by computed means is usually done to find even slight changes that cannot be distinguished via human eye [16]. Analysing the image texture is of vital importance as it allows better plan of action for the clinicians.…”

Section: Biomedical Research 2018; 29 (7): 1316-1320mentioning

confidence: 99%

Influence of gain settings on strain ratios of elastographic image and texture parameters of B-mode image on thyroid tissue

Javed¹,

Baig²,

Mirza³

2018

biomedicalresearch

View full text Add to dashboard Cite

Thyroid nodules are benign though suspicious. Examination through elastography technique is beneficial as it provides real time and first hand evidence about the tissue stiffness by providing a strain value on comparing suspected and normal tissue. Until now, there is no established method that allows to categorise strain ratios whether tumour or normal. Changes in gain settings can significantly alter measurement and ultimately the clinical findings. The objectives of this study are to assess the effect of gain settings on strain ratios of isoechoic stiff lesions in a thyroid gland phantom, and the difference in strain ratios with first order texture parameters at different gain settings. In this study we have reported the effect of gain setting parameter on strain ratios of the thyroid tumour. Furthermore, first order texture parameters (mean and variance) were discussed with respect to increase in gain settings. Higher strain ratios were observed as gain increased. Significant differences in first order texture parameters can be observed for mean of tumour when compared with normal thyroid tissue. Variance did not demonstrate any significant difference. It was concluded that gain parameter effect the strain ratios and texture parameters. This may affect the interpretation of results. It is advised to use elastography in combination with B-mode ultrasound, and sonographers must use a balanced approach when changing equipment parameters.

show abstract

“…Step 3, refine a subset of the features which suits the classification problem; Step 4, use the selected feature set to classify examples based on appropriate supervised or unsupervised classifiers (Vince et al, 2000;Smutek et al, 2001;Lu et al, 2003;Zhu et al, 2006;Chrzanowski et al, 2008). Chen et al(2005) distinguished solid malignant from benign breast lesions using fractal textural features.…”

Section: Ultrasound Image Recognitionmentioning

confidence: 99%

A fast automatic recognition and location algorithm for fetal genital organs in ultrasound images

Tang

Chen

2009

J. Zhejiang Univ. Sci. B

View full text Add to dashboard Cite

Abstract:Severe sex ratio imbalance at birth is now becoming an important issue in several Asian countries. Its leading immediate cause is prenatal sex-selective abortion following illegal sex identification by ultrasound scanning. In this paper, a fast automatic recognition and location algorithm for fetal genital organs is proposed as an effective method to help prevent ultrasound technicians from unethically and illegally identifying the sex of the fetus. This automatic recognition algorithm can be divided into two stages. In the 'rough' stage, a few pixels in the image, which are likely to represent the genital organs, are automatically chosen as points of interest (POIs) according to certain salient characteristics of fetal genital organs. In the 'fine' stage, a specifically supervised learning framework, which fuses an effective feature data preprocessing mechanism into the multiple classifier architecture, is applied to every POI. The basic classifiers in the framework are selected from three widely used classifiers: radial basis function network, backpropagation network, and support vector machine. The classification results of all the POIs are then synthesized to determine whether the fetal genital organ is present in the image, and to locate the genital organ within the positive image. Experiments were designed and carried out based on an image dataset comprising 658 positive images (images with fetal genital organs) and 500 negative images (images without fetal genital organs). The experimental results showed true positive (TP) and true negative (TN) results from 80.5% (265 from 329) and 83.0% (415 from 500) of samples, respectively. The average computation time was 453 ms per image.

show abstract

Image texture analysis of sonograms in chronic inflammations of thyroid gland

Cited by 52 publications

References 28 publications

Systematic Construction of Texture Features for Hashimoto’s Lymphocytic Thyroiditis Recognition from Sonographic Images

Systematic Construction of Texture Features for Hashimoto’s Lymphocytic Thyroiditis Recognition from Sonographic Images

Influence of gain settings on strain ratios of elastographic image and texture parameters of B-mode image on thyroid tissue

A fast automatic recognition and location algorithm for fetal genital organs in ultrasound images

Contact Info

Product

Resources

About