SFFS–SVM based prostate carcinoma diagnosis in DCE-MRI via ACM segmentation
- 42 Downloads
The prostate carcinoma is amongst the most commonly occurring cancers in Taiwanese males. Moreover, it is one of the chief reasons for cancer deaths among Taiwanese men, and early diagnosis of prostate cancer is vital for effective treatment. In this work, a diagnosis model for identifying the prostate carcinoma in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is proposed. The urologists utilize the DCE-MRI as a support mechanism for better diagnosis of the carcinoma development in the prostate. Gadolinium is utilized as the contrast agent for the DCE-MRI data, and it was injected once and the time series data were captured at distinct time intervals of 0, 20, 60, and 100 s correspondingly. Primarily, after pre-processing the DCE-MRI information, the prostate data is segmented by employing the active contour model. Subsequently, 136 features are extracted from the segmented prostrate expanse of the DCE-MRI data, and the relative intensity change curve is computed. Afterward, Fisher’s discriminant ratio and sequential forward floating selection is deployed for choosing ten highly discriminative features. Lastly, the segmented prostate regions are classified into two groups, namely: tumor and normal classes by employing the support vector machine classifier. The experimental results elucidate that the proposed system is superior on the subject of accuracy, sensitivity, and specificity when compared with specific existing methods. Additionally, the proposed system also demonstrates a 94.75% accuracy. Moreover, this signifies the fact that the proposed method for analyzing the DCE data has shown prodigious prospects in the prostate carcinoma diagnosis.
KeywordsProstate carcinoma Dynamic contrast-enhanced MRI Sequential forward floating selection Support vector machine Active contour model
This work was supported by Ministry of Science and Technology, Taiwan, ROC, under Grant MOST 103-2221-E-224-016-MY3.
This research was partially funded by “Intelligent Recognition Industry Service Research Center” from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan (Grant No. 108N04-2) and Ministry of Science and Technology in Taiwan (Grant No. MOST 103-2221-E-224-016-MY3).
- Albayrak, N. B., Betul Oktay, A., & Akgul, Y. S. (2015). Prostate detection from abdominal ultrasound images: A part based approach. In 2015 IEEE international conference on image processing (ICIP), Quebec City, QC (pp. 1955–1959). https://doi.org/10.1109/icip.2015.7351142.
- Artan, Y., Haider, M. A., Langer, D. L., van der Kwast, T. H., Evans, A. J., Yang, Y., et al. (2010). Prostate cancer localization with multispectral MRI using cost-sensitive support vector machines and conditional random fields. IEEE Transactions on Image Processing, 19(9), 2444–2455.MathSciNetzbMATHCrossRefGoogle Scholar
- Azizi, S., Imani, F., Ghavidel, S., Tahmasebi, A., Kwak, J. T., Xu, S., et al. (2016). Detection of prostate cancer using temporal sequences of ultrasound data: A large clinical feasibility study. International Journal of Computer Assisted Radiology and Surgery, 11(6), 947–956.CrossRefGoogle Scholar
- Cancer Society Atlanta. (2011). http://www.cancer.org.
- Castaneda, B., Hoyt, K., Zhang, M., Pasternack, D., Baxter, L., Nigwekar, P., et al. (2007). P1C-9 prostate cancer detection based on three dimensional sonoelastography. In Ultrasonics symposium, 2007. IEEE, New York, NY (pp. 1353–1356). https://doi.org/10.1109/ultsym.2007.340.
- Chang, C.-Y., Hu, H.-Y., & Tsai, Y.-S. (2015). Prostate cancer detection in dynamic MRIs. In 2015 IEEE international conference on digital signal processing (DSP), Singapore (pp. 1279–1282).Google Scholar
- Department of Health and Welfare Taiwan. (2018). 2017 statistics of causes of death in Taiwan.Google Scholar
- Hegde, J. V., Mulkern, R. V., Panych, L. P., Fennessy, F. M., Fedorov, A., Maier, S. E., et al. (2013). Multiparametric MRI of prostate cancer: An update on state-of-the-art techniques and their performance in detecting and localizing prostate cancer. Journal of Magnetic Resonance Imaging, 37(5), 1035–1054.CrossRefGoogle Scholar
- Lavasani, S. N., Mostaar, A., & Ashtiyani, M. (2018). Automatic prostate cancer segmentation using kinetic analysis in dynamic contrast-enhanced MRI. Journal of Biomedical Physics and Engineering, 8(1), 107–116.Google Scholar
- Le Bris, A., Chehata, N., Briottet, X., & Paparoditis, N. (2014). Use intermediate results of wrapper band selection methods: A first step toward the optimization of spectral configuration for land cover classifications. In 2014 6th workshop on hyperspectral image and signal processing: Evolution in remote sensing (WHISPERS). IEEE.Google Scholar
- Lin, J. M. (2012). Prostate cancer segmentation in dynamic MRI. M.S. thesis, National Yunlin University of Science and Technology.Google Scholar
- Ma, Z., Tavares, J. M. R. S., & Jorge, R. N. (2009). A review on the current segmentation algorithms for medical images. In Proceedings of the first international conference on computer imaging theory and applications (Vol. 1, pp. 135–140).Google Scholar
- Moradi, M., Abolmaesumi, P., Isotalo, P. A., Siemens, D. R., Sauerbrei, E. E., & Mousavi, P. (2006). P3E-7 a new feature for detection of prostate cancer based on RF ultrasound echo signals. In 2006 IEEE ultrasonics symposium, Vancouver, BC (pp. 2084–2087). https://doi.org/10.1109/ultsym.2006.530.
- Nixon, M. (2012). Feature extraction and image processing for computer vision. San Diego: Academic Press.Google Scholar
- Pudil, P., Ferri, F. J., Novovicova, J., & Kittler, J. (1994). Floating search methods for feature selection with nonmonotonic criterion functions. In Proceedings of the 12th IAPR international conference on pattern recognition, 1994. Conference B: Computer vision and image processing, Jerusalem (Vol. 2, pp. 279–283). https://doi.org/10.1109/icpr.1994.576920.
- Sammouda, R., Aboalsamh, H., & Saeed, F. (2015). Comparison between K mean and fuzzy C-mean methods for segmentation of near infrared fluorescent image for diagnosing prostate cancer. In 2015 international conference on computer vision and image analysis applications (ICCVIA), Sousse (pp. 1–6). https://doi.org/10.1109/iccvia.2015.7351905.
- Srinivasan, K., & Kanakaraj, J. (2011). A study on super-resolution image reconstruction techniques. Computer Engineering and Intelligent Systems, 2, 222–227.Google Scholar
- Villers, A., Puech, P., Mouton, D., Leroy, X., Ballereau, C., & Lemaitre, L. (2006). Dynamic contrast enhanced, pelvic phased array magnetic resonance imaging of localized prostate cancer for predicting tumor volume: Correlation with radical prostatectomy findings. Journal of Urology, 176, 2432–2437.CrossRefGoogle Scholar
- Zidan, S., & Tantawy, H. I. (2015). Prostate carcinoma: Accuracy of diagnosis and differentiation with dynamic contrast-enhanced MRI and diffusion weighted imaging. The Egyptian Journal of Radiology and Nuclear Medicine, 46(4), 1193–1203. https://doi.org/10.1016/j.ejrnm.2015.06.021.CrossRefGoogle Scholar