2016
DOI: 10.1007/s11760-016-1018-y
|View full text |Cite
|
Sign up to set email alerts
|

Asymmetry analysis of breast thermograms using automated segmentation and texture features

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
20
0
1

Year Published

2018
2018
2022
2022

Publication Types

Select...
5
4

Relationship

1
8

Authors

Journals

citations
Cited by 71 publications
(21 citation statements)
references
References 23 publications
0
20
0
1
Order By: Relevance
“…Breast thermogram images are segmented automatically to separate right and left breast for asymmetry analysis (Sathish et al, 2017). ROI of the breast is segmented from the background by applying Canny edge detection on the image to extract outer edges of the image.…”
Section: Methodsmentioning
confidence: 99%
“…Breast thermogram images are segmented automatically to separate right and left breast for asymmetry analysis (Sathish et al, 2017). ROI of the breast is segmented from the background by applying Canny edge detection on the image to extract outer edges of the image.…”
Section: Methodsmentioning
confidence: 99%
“…Sathish et al [89] showed tracing shape edges by applying a polynomial curve fitting. This method was able to detect slight concavities in the upper border and convexities in the lower part.…”
Section: A Image Pre-processingmentioning
confidence: 99%
“…The resulting accuracy rates were 85%, 80%, and 92.5% for SVM, Naïve Bayes, and k-NN, respectively. A further study [89] classified normal and abnormal breast thermograms based on GLCM textural features and histograms. To improve the accuracy rate, SVM was tested with various kernel functions such as linear, radial basis function (RBF), polynomial, quadratic, and multilayer perceptron.…”
Section: Research On Breast Thermogram Classificationmentioning
confidence: 99%
“…This method used two types of projection, Horizontal Projection Profile (HPP) to detect the upper and lower borders, and Vertical Projection Profile (VPP) to detect the left and right borders. Authors in [44] used a new technique of segmentation after using HPP called asymmetry analysis which depends on finding the point of intersection to separate right and left breasts. In [45], the authors used a technique of segmentation called hot region segmentation method which depends on separating objects from background clearly after applying the technique of k-means clustering which was used to classify colors for Lab mode after the conversion from RGB mode to show the difference between colors and measure that difference.…”
Section: Related Workmentioning
confidence: 99%