Abstract
Mammography is a medical imaging technique which utilizes low-energy X-rays specifically for imaging of breast tissue. Being able to classify mammographic tumors as benign or malignant in the early diagnostic stages can help decrease the amount of subsequent examinations, thus decreasing the associated inappropriate diagnostics (such as false-positives and false-negatives). The following paper describes a method using an artificial neural network in order to predict the severity of a mammographic tumor as either benign or malignant, in order to enhance the overall digital diagnostic procedure of mammographic screening. A feedforward neural network architecture was developed using a data set provided by Institute of Radiology at University Erlangen-Nuremberg. K-fold cross validation was utilized in artificial neural network training, and the effect of varying quantities of neurons in the hidden layer was evaluated by assessing the system output. The single-layer, feedforward neural network architecture created, with 80 neurons in hidden layer, achieved the best performance. The overall structure resulted a sensitivity of 85.0%, specificity of 81.0%, and accuracy of 82.9% in classifying mammographic tumors as benign or malignant.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abraham, A.: Artificial neural networks. In: Sydenham, P., Thorn, R. (eds.) Handbook of Measuring System Design. Wiley, London (2005)
American Cancer Society. What is breast cancer? https://www.cancer.org/cancer/breastcancer/about/what-is-breast-cancer.html (21 Sep 2017). Retrieved 13 Dec 2018
Cochrane Nordic Center. Mammography-leaflet; screening for breast cancer with mammography [pamphlet]. The Nordic Cochrane Center. http://www.cochrane.dk/screening/mammography-leaflet.pdf (2005)
Elter, M., Schulz-Wendtland, R., Wittenberg, T.: The prediction of breast cancer biopsy outcomes using two CAD approaches that both emphasize an intelligible decision process (dissertation, 2007). Med. Phys. 34(11), 4164–4172 (2007)
Lourakis, M.: A brief description of the Levenberg-Marquardt algorithm implemented by Levmar. Doctoral Dissertation, Foundation for Research and Technology—Hellas (2005) [Abstract]. https://www.researchgate.net/publication/239328019_A_Brief_Description_of_the_Levenberg-Marquardt_Algorithm_Implemened_by_levmar (11 Feb 2005). Retrieved 23 Dec 2018
Marquardt, D.: An algorithm for least-squares estimation of nonlinear parameters (doctoral dissertation, 1963). SIAM J. Appl. Math. 11(2), 431–441 (1963). https://doi.org/10.1137/0111030
National Cancer Institute: Mammograms. https://www.cancer.gov/types/breast/mammograms-fact-sheet (7 Dec 2016). Retrieved 13 Dec 2018
National Cancer Institute (Dictionary of Cancer Terms). Definition of spiculated mass (2018). https://www.cancer.gov/publications/dictionaries/cancer-terms/def/spiculatedmass. Retrieved 13 Dec 2018
Rouse, M.: Expert System (Petersen, S. Contr.). https://searchenterpriseai.techtarget.com/definition/expert-system (2018). Retrieved 23 Dec 2018
Sapna, S., Tamilarasi, A., Pravin Kumar, M.: Backpropagation learning algorithm based on Levenberg Marquardt algorithm. Unpublished Doctoral Dissertation. College of Engineering, Tamil Nadu, India
Schulz-Wendtland, R.: Mammographic mass data (Elter, M. Donor). https://archive.ics.uci.edu/ml/machinelearning-databases/mammographic-masses/mammographic_masses.names (Oct 2007). Retrieved 13 Dec 2018
Subaşı, A., Yılmaz, M., Özçelik, H.R.: Classification of EMG signals using wavelet neural network (doctoral dissertation, 2006). J. Neurosci. Methods 156(1–2), 360–367 (2006). doi:10.1016 PMID: 16621003
Šahinbegovic, H., Mušić, L., Alić, B.: Distinguishing physical actions using an artificial neural network. In: XXVI International Conference on Information, Communication and Automation Technologies (ICAT), Sarajevo, Bosnia and Herzegovina, 26–28 Oct 2017
The Mathworks, Inc.: Levenberg-Marquardt backpropagation https://www.mathworks.com/help/deeplearning/ref/trainlm.html;jsessionid=96ec0f3d4f26067224703e6bd3ee (n.d.). Retrieved 23 Dec 2018
World Health Organization (WHO). Diagnostic imaging: mammography https://www.who.int/diagnostic_imaging/imaging_modalities/dim_mammography/en/ (04 Aug 2015). Retrieved 13 Dec 2018
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Mušić, L., Gabeljić, N. (2020). Predicting the Severity of a Mammographic Tumor Using an Artificial Neural Network. In: Badnjevic, A., Škrbić, R., Gurbeta Pokvić, L. (eds) CMBEBIH 2019. CMBEBIH 2019. IFMBE Proceedings, vol 73. Springer, Cham. https://doi.org/10.1007/978-3-030-17971-7_115
Download citation
DOI: https://doi.org/10.1007/978-3-030-17971-7_115
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-17970-0
Online ISBN: 978-3-030-17971-7
eBook Packages: EngineeringEngineering (R0)