ABSTRACT
Considering that Prostate Cancer (PCa) is the most frequently diagnosed tumor in Western men, considerable attention has been devoted in computer-assisted PCa detection approaches. However, this task still represents an open research question. In the clinical practice, multiparametric Magnetic Resonance Imaging (MRI) is becoming the most used modality, aiming at defining biomarkers for PCa. In the latest years, deep learning techniques have boosted the performance in prostate MR image analysis and classification. This work explores the use of the Semantic Learning Machine (SLM) neuroevolution algorithm to replace the backpropagation algorithm commonly used in the last fully-connected layers of Convolutional Neural Networks (CNNs). We analyzed the non-contrast-enhanced multispectral MRI sequences included in the PROSTATEx dataset, namely: T2-weighted, Proton Density weighted, Diffusion Weighted Imaging. The experimental results show that the SLM significantly outperforms XmasNet, a state-of-the-art CNN. In particular, with respect to XmasNet, the SLM achieves higher classification accuracy (without neither pre-training the underlying CNN nor relying on backprogation) as well as a speed-up of one order of magnitude.
- Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, et al. 2015. TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. https://www.tensorflow.org/ Software available from tensorflow.org.Google Scholar
- Samuel G Armato, Henkjan Huisman, Karen Drukker, Lubomir Hadjiiski, Justin S Kirby, Nicholas Petrick, George Redmond, Maryellen L Giger, Kenny Cha, Artem Mamonov, et al. 2018. PROSTATEx Challenges for computerized classification of prostate lesions from multiparametric magnetic resonance images. J. Med. Imaging 5, 4 (2018), 044501.Google ScholarCross Ref
- Léon Bottou. 2010. Large-scale machine learning with stochastic gradient descent. In Proc. COMPSTAT'2010. Springer, 177--186.Google ScholarCross Ref
- Stefano Cagnoni, Andrew B Dobrzeniecki, Riccardo Poli, and Jacquelyn C Yanch. 1999. Genetic algorithm-based interactive segmentation of 3D medical images. Image Vis. Comput. 17, 12 (1999), 881--895.Google ScholarCross Ref
- Young Jun Choi, Jeong Kon Kim, Namkug Kim, Kyoung Won Kim, Eugene K. Choi, and Kyoung-Sik Cho. 2007. Functional MR Imaging of Prostate Cancer. RadioGraphics 27, 1 (jan 2007), 63--75.Google ScholarCross Ref
- François Chollet et al. 2015. Keras. https://keras.io.Google Scholar
- Kenneth Clark, Bruce Vendt, Kirk Smith, John Freymann, Justin Kirby, Paul Koppel, Stephen Moore, Stanley Phillips, David Maffitt, Michael Pringle, et al. 2013. The Cancer Imaging Archive (TCIA): maintaining and operating a public information repository. J. Digit. Imaging 26, 6 (2013), 1045--1057.Google ScholarCross Ref
- Tyler Clark, Junjie Zhang, Sameer Baig, Alexander Wong, Masoom A Haider, and Farzad Khalvati. 2017. Fully automated segmentation of prostate whole gland and transition zone in diffusion-weighted MRI using convolutional neural networks. J. Med. Imaging 4, 4 (2017), 041307.Google ScholarCross Ref
- Anna Fabijańska. 2016. A novel approach for quantification of time-intensity curves in a DCE-MRI image series with an application to prostate cancer. Comput. Biol. Med. 73 (2016), 119--130. Google ScholarDigital Library
- Eleftherios Garyfallidis, Matthew Brett, Bagrat Amirbekian, Ariel Rokem, Stefan Van Der Walt, Maxime Descoteaux, and Ian Nimmo-Smith. 2014. DIPy, a library for the analysis of diffusion MRI data. Front. Neuroinform. 8 (2014), 8.Google ScholarCross Ref
- Ivo Gonçalves, Sara Silva, and Carlos M Fonseca. 2015. On the generalization ability of geometric semantic genetic programming. In Genetic Programming. Springer, 41--52.Google Scholar
- Ivo Gonçalves, Sara Silva, and Carlos M. Fonseca. 2015. Semantic learning machine: a feedforward neural network construction algorithm inspired by geometric semantic genetic programming. In Progress in Artificial Intelligence. Lecture Notes in Computer Science, Vol. 9273. Springer, 280--285.Google Scholar
- Ivo Gonçalves, Sara Silva, Carlos M Fonseca, and Mauro Castelli. 2016. Arbitrarily close alignments in the error space: a geometric semantic genetic programming approach. In Proceedings of the 2016 on Genetic and Evolutionary Computation Conference Companion. ACM, 99--100. Google ScholarDigital Library
- Ivo Gonçalves. 2017. An Exploration of Generalization and Overfitting in Genetic Programming: Standard and Geometric Semantic Approaches. Ph.D. Dissertation. Department of Informatics Engineering, University of Coimbra, Portugal.Google Scholar
- Ivo Gonçalves, Sara Silva, Carlos M. Fonseca, and Mauro Castelli. 2017. Unsure when to Stop? Ask Your Semantic Neighbors. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO). ACM, New York, NY, USA, 929--936. Google ScholarDigital Library
- Vikas Gulani, Fernando Calamante, Frank G Shellock, Emanuel Kanal, Scott B Reeder, et al. 2017. Gadolinium deposition in the brain: summary of evidence and recommendations. Lancet Neurol. 16, 7 (2017), 564--570.Google ScholarCross Ref
- Yanrong Guo, Yaozong Gao, and Dinggang Shen. 2016. Deformable MR prostate segmentation via deep feature learning and sparse patch matching. IEEE Trans. Med. Imaging 35, 4 (2016), 1077--1089.Google ScholarCross Ref
- Masoom A Haider, Theodorus H Van Der Kwast, Jeff Tanguay, Andrew J Evans, Ali-Tahir Hashmi, Gina Lockwood, and John Trachtenberg. 2007. Combined T2-weighted and diffusion-weighted MRI for localization of prostate cancer. AJR Am. J. Roentgenol. 189, 2 (2007), 323--328.Google ScholarCross Ref
- Thomas Hambrock, Pieter C Vos, Christina A Hulsbergen-van de Kaa, Jelle O Barentsz, and Henkjan J Huisman. 2013. Prostate cancer: computer-aided diagnosis with multiparametric 3-T MR imaging-effect on observer performance. Radiology 266, 2 (2013), 521--530.Google ScholarCross Ref
- Jan-Benedikt Jagusch, Ivo Gonçalves, and Mauro Castelli. 2018. Neuroevolution under unimodal error landscapes: an exploration of the semantic learning machine algorithm.In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO) Companion. ACM, New York, NY, USA, 159--160. Google ScholarDigital Library
- Haozhe Jia, Yong Xia, Yang Song, Weidong Cai, Michael Fulham, and David Dagan Feng. 2018. Atlas registration and ensemble deep convolutional neural network-based prostate segmentation using magnetic resonance imaging. Neurocomputing 275 (2018), 1358--1369. Google ScholarDigital Library
- Daniel Junker, Fabian Steinkohl, Veronika Fritz, Jasmin Bektic, Theodoros Tokas, Friedrich Aigner, Thomas RW Herrmann, Michael Rieger, and Udo Nagele. 2018. Comparison of multiparametric and biparametric MRI of the prostate: are gadolinium-based contrast agents needed for routine examinations? World J. Urol. (2018), 1--9.Google Scholar
- Konstantinos Kamnitsas, Christian Ledig, V. F. J. Newcombe, Joanna P. Simpson, Andrew D. Kane, David K. Menon, et al. 2017. Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med. Image Anal. 36 (2017), 61--78.Google ScholarCross Ref
- Diederik P Kingma and Jimmy Ba. 2014. Adam: a method for stochastic optimization.arXiv preprint arXiv:1412.6980 (2014).Google Scholar
- Deukwoo Kwon, Isildinha M Reis, Adrian L Breto, Yohann Tschudi, Nicole Gautney, Olmo Zavala-Romero, Christopher Lopez, John C Ford, Sanoj Pun-nen, Alan Pollack, et al. 2018. Classification of suspicious lesions on prostate multiparametric MRI using machine learning. J. Med. Imaging 5, 3 (2018), 034502.Google Scholar
- Guillaume Lemaître, Robert Martí, Jordi Freixenet, Joan C Vilanova, Paul M Walker, and Fabrice Meriaudeau. 2015. Computer-aided detection and diagnosis for prostate cancer based on mono and multi-parametric MRI: a review. Comput. Biol. Med. 60 (2015), 8--31. Google ScholarDigital Library
- Roger Li, Gregory C Ravizzini, Michael A Gorin, Tobias Maurer, Matthias Eiber, Matthew R Cooperberg, Mehrdad Alemozzaffar, Matthew K Tollefson, Scott E Delacroix, and Brian F Chapin. 2018. The use of PET/CT in prostate cancer. Prostate Cancer Prostatic Dis. 21, 1 (2018), 4.Google ScholarCross Ref
- Geert Litjens, Oscar Debats, Jelle Barentsz, Nico Karssemeijer, and Henkjan Huisman. 2014. Computer-aided detection of prostate cancer in MRI IEEE Trans. Med. Imaging 33, 5 (2014), 1083--1092.Google ScholarCross Ref
- Geert Litjens, Oscar Debats, Jelle Barentsz, Nico Karssemeijer, and Henkjan Huisman. 2017. "PROSTATEx Challenge data", The Cancer Imaging Archive. https://wiki.cancerimagingarchive.net/display/Public/SPIE-AAPM-NCI+PROSTATEx+Challenges. Online; Accessed on January 25, 2019.Google Scholar
- Saifeng Liu, Huaixiu Zheng, Yesu Feng, and Wei Li. 2017. Prostate cancer diagnosis using deep learning with 3D multiparametric MRI. In Medical Imaging 2017: Computer-Aided Diagnosis (Proceedings SPIE), Vol. 10134. International Society for Optics and Photonics, 1013428.Google Scholar
- F. Maes, A. Collignon, D. Vandermeulen, G. Marchal, and P. Suetens. 1997. Multi-modality image registration by maximization of mutual information. IEEE Trans. Med. Imaging 16, 2 (apr 1997), 187--198.Google ScholarCross Ref
- Lukasz Matulewicz, Jacobus FA Jansen, Louisa Bokacheva, Hebert Alberto Vargas, Oguz Akin, Samson W Fine, Amita Shukla-Dave, James A Eastham, Hedvig Hricak, Jason A Koutcher, et al. 2014. Anatomic segmentation improves prostate cancer detection with artificial neural networks analysis of 1H magnetic resonance spectroscopic imaging. J. Magn. Reson. Imaging 40, 6 (2014), 1414--1421.Google ScholarCross Ref
- Fausto Milletari, Nassir Navab, and Seyed-Ahmad Ahmadi. 2016. V-Net: fully convolutional neural networks for volumetric medical image segmentation.In Proceedings of the International Conference on 3D Vision (3DV). IEEE, 565--571.Google Scholar
- Alberto Moraglio, Krzysztof Krawiec, and Colin G Johnson. 2012. Geometric semantic genetic, programming. In Parallel Problem Solving from Nature-PPSN XII. Springer, 21--31. Google ScholarDigital Library
- Alberto Moraglio and Andrea Mambrini. 2013. Runtime analysis of mutation-based geometric semantic genetic programming for basis functions regression. In Proceedings of the 15th Annual Conference on Genetic and Evolutionary Computation (GECCO). ACM, 989--996. Google ScholarDigital Library
- Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-Net: convolutional Networks for Biomedical Image Segmentation. In Proceedings of the Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) (LNCS), Vol. 9351. Springer, 234--241.Google ScholarCross Ref
- Andrew B Rosenkrantz, Ruth P Lim, Mershad Haghighi, Molly B Somberg, James S Babb, and Samir S Taneja. 2013. Comparison of interreader reproducibility of the prostate imaging reporting and data system and Likert scales for evaluation of multiparametric prostate MRI. AJR Am. J. Roentgenol. 201, 4 (2013), W612-W618.Google ScholarCross Ref
- Leonardo Rundo, Carmelo Militello, Giorgio Russo, Antonio Garufi, Salvatore Vitabile, Maria Carla Gilardi, and Giancarlo Mauri. 2017. Automated prostate gland segmentation based on an unsupervised fuzzy c-means clustering technique using multispectral T1w and T2w MR imaging. Information 8, 2 (2017), 49.Google ScholarCross Ref
- L. Rundo, A. Stefano, C. Militello, G. Russo, M. G. Sabini, C. D'Arrigo, F. Marletta, M. Ippolito, G. Mauri, S. Vitabile, and M. C. Gilardi. 2017. A fully automatic approach for multimodal PET and MR image segmentation in Gamma Knife treatment planning. Comput. Methods Programs Biomed. 144 (2017), 77--96. Google ScholarDigital Library
- Leonardo Rundo, Andrea Tangherloni, Paolo Cazzaniga, Marco S Nobile, Giorgio Russo, Maria Carla Gilardi, Salvatore Vitabile, Giancarlo Mauri, Daniela Besozzi, and Carmelo Militello. 2019. A novel framework for MR image segmentation and quantification by using MedGA. Comput. Methods Programs Biomed. (2019). In press.Google Scholar
- Leonardo Rundo, Andrea Tangherloni, Carmelo Militello, Maria Carla Gilardi, and Giancarlo Mauri. 2016. Multimodal medical image registration using particle swarm optimization: a review. In Proc. Symposium Series on Computational Intelligence (SSCI). IEEE, 1--8.Google ScholarCross Ref
- Leonardo Rundo, Aandrea Tangherloni, Marco S. Nobile, Carmelo Militello, Daniela Besozzi, Giancarlo Mauri, and Paolo Cazzaniga. 2019. MedGA: a novel evolutionary method for image enhancement in medical imaging systems. Expert Syst. Appl. 119 (2019), 387--399.Google ScholarCross Ref
- Angela Serra, Paola Galdi, and Roberto Tagliaferri. 2018. Machine learning for bioinformatics and neuroimaging. Wiley Interdisc. Rev. Data Min. Knowl. Discov. 8, 5 (2018), e1248.Google ScholarCross Ref
- Rebecca L Siegel, Kimberly D Miller, and Ahmedin Jemal. 2019. Cancer statistics, 2019. CA Cancer J. Clin. 69, 1 (2019), 7--34.Google ScholarCross Ref
- Karen Simonyan and Andrew Zisserman. 2015. Very deep convolutional networks for large-scale image recognition. In Proceedings of the International Conference on Learning Representations (ICLR). arXiv preprint arXiv:1409.1556.Google Scholar
- Jost Tobias Springenberg, Alexey Dosovitskiy, Thomas Brox, and Martin Ried-miller. 2014. Striving for simplicity: The all convolutional net. In Proceedings of the International Conference on Learning Representations (ICLR). 1--14. arXiv preprint arXiv:1412.6806.Google Scholar
- Radka Stoyanova, Mandeep Takhar, Yohann Tschudi, John C Ford, Gabriel Solórzano, Nicholas Erho, Yoganand Balagurunathan, Sanoj Punnen, Elai Davi-cioni, Robert J Gillies, et al. 2016. Prostate cancer radiomics and the promise of radiogenomics. Transl. Cancer Res. 5, 4 (2016), 432.Google ScholarCross Ref
- Yohan Sumathipala, Nathan Lay, Baris Turkbey, Clayton Smith, Peter L Choyke, and Ronald M Summers. 2018. Prostate cancer detection from multi-institution multiparametric MRIs using deep convolutional neural networks. J. Med. Imaging 5, 4 (2018), 044507.Google ScholarCross Ref
- Jinquan Sun, Yinghuan Shi, Yang Gao, and Dinggang Shen. 2017. A point says a lot: an interactive segmentation method for MR prostate via one-point labeling. In Proceendigs of the International Workshop on Machine Learning in Medical Imaging (MLMI). Springer, 220--228.Google ScholarCross Ref
- Baris Turkbey, Anna M Brown, Sandeep Sankineni, Bradford J Wood, Peter A Pinto, and Peter L Choyke. 2016. Multiparametric prostate magnetic resonance imaging in the evaluation of prostate cancer. CA: A Cancer Journal for Clinicians 66, 4 (2016), 326--336.Google ScholarCross Ref
- Zhiwei Wang, Chaoyue Liu, Danpeng Cheng, Liang Wang, Xin Yang, and Kwang-Ting Cheng. 2018. Automated detection of clinically significant prostate cancer in mp-MRI images based on an end-to-end deep neural network. IEEE Trans. Med. Imaging 37, 5 (2018), 1127--1139.Google ScholarCross Ref
- David H Wolpert and William G Macready. 1997. No free lunch theorems for optimization. IEEE Trans. Evol. Computat. 1, 1 (1997), 67--82. Google ScholarDigital Library
Index Terms
- Semantic learning machine improves the CNN-Based detection of prostate cancer in non-contrast-enhanced MRI
Recommendations
Enhancing classification performance of convolutional neural networks for prostate cancer detection on magnetic resonance images: a study with the semantic learning machine
GECCO '19: Proceedings of the Genetic and Evolutionary Computation Conference CompanionProstate cancer (PCa) is the most common oncological disease in Western men. Even though a significant effort has been carried out by the scientific community, accurate and reliable automated PCa detection methods are still a compelling issue. In this ...
A PET/CT directed, 3D ultrasound-guided biopsy system for prostate cancer
MICCAI'11: Proceedings of the 2011 international conference on Prostate cancer imaging: image analysis and image-guided interventionsProstate cancer affects 1 in 6 men in the USA. Systematic transrectal ultrasound (TRUS)-guided biopsy is the standard method for a definitive diagnosis of prostate cancer. However, this "blind" biopsy approach can miss at least 20% of prostate cancers. ...
TRUS/MRI Fusion-Guided Prostate Biopsy Based on Improved Intracavitary Markers
ICBCI 2017: Proceedings of the International Conference on Bioinformatics and Computational IntelligenceGuiding prostate biopsy with the fusion of magnetic resonance imaging (MRI) and transrectal ultrasonography (TRUS) could improve the diagnostic rate of prostate cancer. Matching and registering the images of the prostate in MRI and TRUS are the two main ...
Comments