A restricted Boltzmann machine based prostate tumor detection in MRI images
Keywords:Prostate Cancer, Magnetic Resonance Imaging, Restricted Boltzmann Machine, Deep Belief Network
Context: Prostate cancer is a common and malignant tumor. Thus, to improve the excellence of life of cancer person is the early diagnosis of prostate cancer. Prostate tumors are divided in terms of growth to: slow growth and often confined to the prostate gland and growing growth and often spread the prostate gland to other members which is the second most regular reason for malignancy in the world According to estimates and studies conducted by the World Health Organization. MRI uses medical imaging technology used on a scale It is widely evaluated for tumors, but the large amount of data produced by MRI can vary greatly. Thus manual detection will be a challenge.
The Problem: The early diagnosis of prostate cancer plays a key role in prolonging the patient's life. as doctors are still human, and rely on factors such as the eye in the diagnosis and these factors do not avoid mistakes in addition to that The wrong diagnosis presents the patient to death, especially the description of inappropriate treatment or surgical intervention (tumor contact) or chemical radiation therapy for this reason completed the diagnosis leads to high accuracy diagnosis, accuracy Diagnosis assumes an imperative job in the fight against the disease. Regular re-diagnosis is important and necessary to ensure that patients survive and keep them away from the risk of disease. However, re-diagnosis and examination often consumes many financial expenses in this regard. This system is designed to assist radiologists in their practice Clinical.
Approach: In this paper, the researcher is interested in focusing on the area of the prostate gland and detecting abnormal cases in the image. The proposed strategy relies on one of the deep learning algorithms Restricted Boltzmann Machine algorithm, and image processing techniques histogram equalization one of Contrast enhancement techniques, to enhance grayscale images It improves the differentiation of pictures by changing the qualities in a force picture so the histogram of the output image approximately matches a specified histogram and extract the features that help us to make the right decision in the process of diagnosis as well as to increase the efficiency of the system and obtain the highest accuracy in the results where the use of dataset images of the magnetic resonance of the prostate, which contains more than 1700 different images of 230 infected and was classified using an algorithm to non-tumor and tumor.
Finding: The results of this study confirmed that this method works effectively. This method was applied to a database containing 1730 medical images. The accuracy of this method was 98.8439. To demonstrate the productivity of the proposed system, the images were entered into a normal neural network. Than Indicates the efficiency of our proposed system.
 Banerjee, S., & Kaviani, A. (2016). Worldwide prostate cancer epidemiology: Differences between regions, races, and awareness programs.â€
 Center, M. M., Jemal, A., Lortet-Tieulent, J., Ward, E., Ferlay, J., Brawley, O., & Bray, F. (2012). International variation in prostate cancer incidence and mortality rates. European urology, 61(6), 1079-1092.â€ https://doi.org/10.1016/j.eururo.2012.02.054.
 Lemaitre G., MartÃ R., et al. â€œComputer-Aided Detection for Prostate Cancer Detection based on Multi-Parametric Magnetic Resonance Imagingâ€ EMBC2017: 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Jul 2017, Jeju Island, South Korea.
 Thahseen P, Anish Kumar B, â€œA Deep Belief Network Based Brain Tumor Detection in MRI Imagesâ€, International Journal of Science and Research (IJSR), Vol. 6,2015, pp. 495 ,498.
 Radiological Society of North America, â€œMagnetic Resonance Imaging (MRI) â€“ Bodyâ€, June 2018, http://www.radiologyinfo.org.
 Ampeliotis, D., Antonakoudi, A., Berberidis, K., Psarakis, E. Z., & Kounoudes, A. (2008, March). A computer-aided system for the detection of prostate cancer based on magnetic resonance image analysis. In Communications, Control and Signal Processing, 2008. ISCCSP 2008. 3rd International Symposium on (pp. 1372-1377). IEEE.â€ https://doi.org/10.1109/ISCCSP.2008.4537440.
 Hu, X., Cammann, H., Meyer, H. A., Miller, K., Jung, K., & Stephan, C. (2013). Artificial neural networks and prostate cancerâ€”tools for diagnosis and management. Nature Reviews Urology, 10(3), 174.â€ https://doi.org/10.1038/nrurol.2013.9.
 Zhu, Y., Wang, L., Liu, M., Qian, C., Yousuf, A., Oto, A., & Shen, D. (2017). MRIâ€based prostate cancer detection with highâ€level representation and hierarchical classification. Medical physics, 44(3), 1028-1039.â€ https://doi.org/10.1002/mp.12116.
 Lemaitre, G. (2016). Computer-aided diagnosis for prostate cancer using multi-parametric magnetic resonance imaging (Doctoral dissertation, Ph. D. dissertation, Universitat de Girona and UniversitÃ© de Bourgogne).â€
 Mehrtash, A., Sedghi, A., Ghafoorian, M., Taghipour, M., Tempany, C. M., Wells, W. M., ... & Fedorov, A. (2017, March). Classification of clinical significance of MRI prostate findings using 3D convolutional neural networks. In Medical Imaging 2017: Computer-Aided Diagnosis (Vol. 10134, p. 101342A). International Society for Optics and Photonics.â€
 Firjani, A., Khalifa, F., Elnakib, A., Gimel'farb, G., El-Ghar, M. A., Elmaghraby, A., & El-Baz, A. (2012, September). A novel image-based approach for early detection of prostate cancer. In Image Processing (ICIP), 2012 19th IEEE International Conference on (pp. 2849-2852). IEEE.â€ https://doi.org/10.1109/ICIP.2012.6467493.
 The Ferenc Jolesz National Center for Image Guided Therapy, Harvard medical school, Brigham and woman school,â€ Prostate MR Image Databaseâ€, 2008, http://prostatemrimagedatabase.com/index.html.
 Musthofa Sunaryo et al, â€œPreprocessing on Digital Image using Histogram Equalization: An Experiment Study on MRI Brain Image / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 7 (4) 2016, 1723-1727.
 Hebbo H., Won Kim J, â€œClassification with Deep Belief Networksâ€.
 Hinton, G. E. (2012). A practical guide to training restricted Boltzmann machines. In Neural networks: Tricks of the trade (pp. 599-619). Springer, Berlin, Heidelberg https://doi.org/10.1007/978-3-642-35289-8_32.
 Fischer A., Igel CH., â€œAn Introduction to Restricted Boltzmann Machinesâ€, (Eds.): CIARP 2012, LNCS 7441, pp. 14â€“36, 2012 https://doi.org/10.1007/978-3-642-33275-3_2.
 Sims, James Christopher, "An Implementation of Deep Belief Networks Using Restricted Boltzmann Machines in Clojure" (2016). Open Access Master's Theses. Paper 804. http://digitalcommons.uri.edu/theses/804.
 Sebastian V B., Unnikrishnan A, et al.â€ Grey Level Co-occurrence Matrices: Generalisation and some new features â€œ, International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol.2, No.2, April 2012.
 Pathak B, Barooah D, â€œTexture Analysis Based on the Gray-Level Co-Occurrence Matrix Considering Possible Orientations, International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering,Vol. 2, September, 2013.
 Grakari DH, â€œImage Quality Analysis Using GLCM MSc. Thesis, University of Central Florida, Orlando, Florida, 2004.
 Sharma, E. K., Priyanka, E., Kalsh, E. A., & Saini, E. K. GLCM and its Features.
 Song, F., Guo, Z., & Mei, D. (2010, November). Feature selection using principal component analysis. In System science, engineering design and manufacturing informatization (ICSEM), 2010 international conference on (Vol. 1, pp. 27-30). IEEE.â€ https://doi.org/10.1109/ICSEM.2010.14.
 Yang, L. (2015). An application of principal component analysis to stock portfolio management.â€
 Chilali, O., Ouzzane, A., Diaf, M., & Betrouni, N. (2014). A survey of prostate modeling for image analysis. Computers in biology and medicine, 53, 190-202.â€ https://doi.org/10.1016/j.compbiomed.2014.07.019.
 Haifeng Wang. "Ensemble learning approach for classification variance reduction in breast cancer diagnosisâ€. MSc. Thesis, Binghamton University, State University of New York, ProQuest LLC. 2015.