Thyroid Texture Representation via Noise Resistant Image Features

Keramidas, Eystratios G.; Iakovidis, Dimitris K.; Maroulis, Dimitris; Dimitropoulos, Nikos

doi:10.1109/cbms.2008.108

Cited by 37 publications

(18 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Then the histogram of the labels can be used as a texture descriptor. The textural characteristics of the thyroid tissue are encoded by histograms of Local Binary Patterns (LBP) [17]. The advantages of this novel scheme include increased accuracy in nodule detection, compared with the conventional ultrasound image analysis methods, and time efficiency.…”

Section: ) Linear Binary Patternmentioning

confidence: 99%

“…Five features short runs emphasis, long runs emphasis, gray level uniformity, run length uniformity, and run percentage are calculated from the gray level run-length matrix. In [12,17] a set of novel fuzzy features for thyroid US texture representation, demonstrating noise-resistant properties are presented, analyzed and evaluated. The textural feature extraction scheme is based on the fuzzification of the LBP approach.…”

Section: ) Linear Binary Patternmentioning

confidence: 99%

See 1 more Smart Citation

Computer-Aided Diagnosis of Thyroid Nodule: A Review

Koundal¹,

Gupta²,

Singh³

2012

IJCSES

View full text Add to dashboard Cite

show abstract

Section: ) Linear Binary Patternmentioning

confidence: 99%

Section: ) Linear Binary Patternmentioning

confidence: 99%

Computer-Aided Diagnosis of Thyroid Nodule: A Review

Koundal¹,

Gupta²,

Singh³

2012

IJCSES

View full text Add to dashboard Cite

show abstract

“…The contour is irregular and asymmetric. A good diagnosis can be done by using ABCDE system (Asymmetrical, Border, Color, Diameter and Evolution), [5]. This means that the temperature map has a higher spatial-temperature resolution as the tumor is smaller and can assure early cancer detection.…”

Section: Introductionmentioning

confidence: 99%

Infrared signature analysis of the thyroid tumors

2009

View full text Add to dashboard Cite

Cancer is a leading cause of death worldwide, and about 30% of cancer deaths can be prevented. In the next future, the number of global cancer deaths is projected to increase 45% in the future. A general treatment has not yet been found. The best defense against cancer is early detection, when tumor dimensions are very small. The methods as mammography, ultrasounds, MRI, CT, etc., can detect anatomic or structural changes like tumors and cysts. They are anatomical imaging procedures, consequently, they have the ability to locate the area of the tumor, but they cannot detect a fast-growing cancer in the pre-invasive stage. Thermograms are looking for the physiologic changes in tissue; which may indicate a risk of developing cancer in the future. The results using a new device, operating in infrared band, are described. The paper focuses on thyroid cancer because it allows investigations on larger areas before surgery and on residual, smaller areas following surgery. The experiment results for 24 patients with thyroid nodules are described. Malign tumors have a distinct infrared signature. Only the area affected is thermal registered and that has an irregular shape and a strong nonuniform structure with rapid variations on skin temperature.

show abstract

“…It has been extensively studied in a wide array of fields and has demonstrated superior performance in several comparative studies [12,13]. In medicine, LBP has been used to identify malignant breast cells [10], to find relevant slices in brain MR (magnetic resonance) volumes [14], and as a texture feature extracted from images of thyroid slices [15]. Recent work has also investigated LBP in automated cell phenotype image classification [16].…”

mentioning

confidence: 99%