Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A deep-wavelet neural network to detect and classify lesions in mammographic images

A deep-wavelet neural network to detect and classify lesions in mammographic images PurposeBreast cancer is still one of the deadliest forms of cancer for women, both in developed countries and in underdeveloped and developing nations. Mammograms are currently the most validated imaging techniques to support the differential diagnosis of malignant and benign lesions. Radiologists often only need to clarify doubts about regions of interest that correspond to suspected lesions. Deep-wavelet neural networks are convolutional neural networks that do not necessarily learn, as they can have predefined filter banks as their neurons.MethodsIn this work, we propose a deep hybrid architecture to support digital mammography region-of-interest imaging diagnosis based on six-layer deep-wavelet neural networks, to extract attributes of regions of interest from mammograms, and support vector machine with kernel second-degree polynomial for final classification.ResultsClassical 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 94% and an average kappa of 0.91, employing a 6-layer deep-wavelet network and a two-degree polynomial kernel support vector machine as the final classifier.ConclusionUsing a deep neural network with prefixed weights from the wavelets transform filter bank, 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 support vector machines with second-degree polynomial kernel. This shows that deep networks that do not learn can be important in building complete solutions to support mammographic imaging diagnosis. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Research on Biomedical Engineering Springer Journals

A deep-wavelet neural network to detect and classify lesions in mammographic images

Download PDF
16 pages

Loading next page...
 
/lp/springer-journals/a-deep-wavelet-neural-network-to-detect-and-classify-lesions-in-T5csBCOtJE

References (73)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to The Brazilian Society of Biomedical Engineering 2022. Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
ISSN
2446-4732
eISSN
2446-4740
DOI
10.1007/s42600-022-00238-8
Publisher site
See Article on Publisher Site

Abstract

PurposeBreast cancer is still one of the deadliest forms of cancer for women, both in developed countries and in underdeveloped and developing nations. Mammograms are currently the most validated imaging techniques to support the differential diagnosis of malignant and benign lesions. Radiologists often only need to clarify doubts about regions of interest that correspond to suspected lesions. Deep-wavelet neural networks are convolutional neural networks that do not necessarily learn, as they can have predefined filter banks as their neurons.MethodsIn this work, we propose a deep hybrid architecture to support digital mammography region-of-interest imaging diagnosis based on six-layer deep-wavelet neural networks, to extract attributes of regions of interest from mammograms, and support vector machine with kernel second-degree polynomial for final classification.ResultsClassical 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 94% and an average kappa of 0.91, employing a 6-layer deep-wavelet network and a two-degree polynomial kernel support vector machine as the final classifier.ConclusionUsing a deep neural network with prefixed weights from the wavelets transform filter bank, 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 support vector machines with second-degree polynomial kernel. This shows that deep networks that do not learn can be important in building complete solutions to support mammographic imaging diagnosis.

Journal

Research on Biomedical Engineering – Springer Journals

Published: Dec 1, 2022

Keywords: Breast cancer; Digital mammography; Breast cancer image diagnosis; Deep learning; Deep wavelets

There are no references for this article.