Feature Extraction from Contours Shape for Tumor Analyzing in Mammographic Images

Boujelben, Atef; Chaabani, Ali; Tmar, Hedi; Abid, M.

doi:10.1109/dicta.2009.71

Cited by 17 publications

(17 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These results are not the best result compared to local works (Boujelben et al, 2009a) (Boujelben et al, 2009b). However, in (Boujelben et al, 2009b) the result is about 94% in boundary information; and in (Boujelben et al, 2009a) the result is between 90% and 92% using XRDM method.…”

Section: Experimental Results: Performance In Terms Of Diagnosis Qualitymentioning

confidence: 61%

“…However, in (Boujelben et al, 2009b) the result is about 94% in boundary information; and in (Boujelben et al, 2009a) the result is between 90% and 92% using XRDM method. In such work, we used DDSM database but the ROI is selected from the image by fixing a rectangular box around the suspicious lesion area and the classical method of segmentation based on Sobel filter and thresholding.…”

Section: Experimental Results: Performance In Terms Of Diagnosis Qualitymentioning

confidence: 99%

“…However, the calculus is limited to the convex points that forms the perimeter of the region with preserving the result quality. Quite close from , we propose a novel measure denoted Index Angle (IA) (Boujelben et al, 2009b). Its based on the external and internal angle concepts.…”

Section: Context Of State Of the Artmentioning

confidence: 99%

“…It is used to describe the regularity of the mammogram masses. However, simple morphologic features like Circularity and Eccentricity are extracted through this descriptor (Yang et al, 2005) (Boujelben et al, 2009b) (Delogu et al, 2008). Alvarenga et al (Alvarenga et al, 2006) evaluated the performance and relevance of seven shape features; namely Perimeter (P), Normalized Radial Length (NRL), SDEV, Area Ratio (AR), contour roughness (R), Circularity and Mshape.…”

Section: Context Of State Of the Artmentioning

confidence: 99%

“…In this context, the manual segmentation and semi-automatic method are widely used. In (Boujelben et al, 2009a) (Boujelben et al, 2009b), the threshold-based segmentation is carried out by fixing a rectangular box around the suspicious tumour area and then using Sobel filter in order to avoid noise. However, this can affect noise and discontinuities in the border of ROI.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Automatic Diagnosis of Breast Tissue

Boujelben¹,

Tmar²,

Abid³

et al. 2012

Advances in Cancer Management

Self Cite

View full text Add to dashboard Cite

Section: Experimental Results: Performance In Terms Of Diagnosis Qualitymentioning

confidence: 61%

Section: Experimental Results: Performance In Terms Of Diagnosis Qualitymentioning

confidence: 99%

Section: Context Of State Of the Artmentioning

confidence: 99%

Section: Context Of State Of the Artmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Automatic Diagnosis of Breast Tissue

Boujelben¹,

Tmar²,

Abid³

et al. 2012

Advances in Cancer Management

Self Cite

View full text Add to dashboard Cite

Artificial Intelligence in Breast Cancer Diagnosis: A Review

Karampotsis,

Panourgias,

Dounias

2024

Learning and Analytics in Intelligent Systems

View full text Add to dashboard Cite

Combining deep-wavelet neural networks and support-vector machines to classify breast lesions in thermography images

et al. 2022

View full text Add to dashboard Cite

Breast cancer is among the leading causes of cancer death among women. The occurrence of breast cancer is similar both in developed countries and in underdeveloped and developing nations, although mortality is higher in underdeveloped countries due to late detection. Even though mammography is the most used technique for the differential diagnosis of breast cancer, breast thermography can be used as a complementary technique, more accurate than self-examination but accurate enough to guide the use of a mammogram. However, thermal imaging is still difficult for radiologists to understand. Machine learning can help improve this scenario. Deep Wavelet Neural Networks are convolutional neural networks that do not necessarily learn, as they may have predefined filter banks as their neurons. Methods: In this work, we propose a deep hybrid architecture to support breast thermography imaging diagnosis based on five-layer Deep-Wavelet * Corresponding author: Wellington Pinheiro dos Santos, ORCID 0000-0003-2558-6602 DWNN to classify breast lesions in thermographies Neural Networks, to extract attributes of regions of interest from mammograms, and linear kernel support vector machines for final classification. Results: Classical classifiers such as Bayesian classifiers, single hidden layer multilayer perceptrons, decision trees, Random Forests, and support vector machines were tested. The results showed that it is possible to detect and classify injuries with an average accuracy of 99% and an average kappa of 0.99, employing a 5-layer deep-wavelet network and a linear kernel support vector machine as the final classifier. Conclusion: Using a deep neural network with prefixed weights from the Wavelets Transform filterbank, it was possible to extract attributes and thus take the problem to a universe where it can be solved with relatively simple decision boundaries like those composed by linear kernel support vector machines. This shows that these new deep networks can be important in building complete solutions to improve breast thermography imaging results to support clinical diagnosis.

show abstract

Feature Extraction from Contours Shape for Tumor Analyzing in Mammographic Images

Cited by 17 publications

References 6 publications

Automatic Diagnosis of Breast Tissue

Automatic Diagnosis of Breast Tissue

Artificial Intelligence in Breast Cancer Diagnosis: A Review

Combining deep-wavelet neural networks and support-vector machines to classify breast lesions in thermography images

Contact Info

Product

Resources

About