Predicting prostate tumour location from multiparametric MRI using Gaussian kernel support vector machines: a preliminary study
- 492 Downloads
The performance of a support vector machine (SVM) algorithm was investigated to predict prostate tumour location using multi-parametric MRI (mpMRI) data. The purpose was to obtain information of prostate tumour location for the implementation of bio-focused radiotherapy. In vivo mpMRI data were collected from 16 patients prior to radical prostatectomy. Sequences included T2-weighted imaging, diffusion-weighted imaging and dynamic contrast enhanced imaging. In vivo mpMRI was registered with ‘ground truth’ histology, using ex vivo MRI as an intermediate registration step to improve accuracy. Prostate contours were delineated by a radiation oncologist and tumours were annotated on histology by a pathologist. Five patients with minimal imaging artefacts were selected for this study. A Gaussian kernel SVM was trained and tested on different patient data subsets. Parameters were optimised using leave-oneout cross validation. Signal intensities of mpMRI were used as features and histology annotations as true labels. Prediction accuracy, as well as area under the curve (AUC) of the receiver operating characteristics (ROC) curve, were used to assess performance. Results demonstrated the prediction accuracy ranged from 70.4 to 87.1% and AUC of ROC ranged from 0.81 to 0.94. Additional investigations showed the apparent diffusion coefficient map from diffusion weighted imaging was the most important imaging modality for predicting tumour location. Future work will incorporate additional patient data into the framework to increase the sensitivity and specificity of the model, and will be extended to incorporate predictions of biological characteristics of the tumour which will be used in bio-focused radiotherapy optimisation.
KeywordsProstate cancer Multiparametric MRI Machine learning Support vector machines Focal therapy Bio-focused therapy
This research is supported by the Prostate Cancer Foundation of Australia, The University of Melbourne and Cancer Therapeutics CRC. The authors would also like to show their gratitude to Courtney Savill and Lauren Caspersz who have made substantial contributions during data collections.
Compliance with Ethical Standards
Conflict of interest
The authors declare that they have no conflicts of interest.
Ethics in this article is approved by Human Research Ethics Committee (HREC).
- 1.AIHW (2013) Prostate cancer in Australia. AIHW, CanberraGoogle Scholar
- 7.Giannini V et al (2013) A prostate CAD system based on multiparametric analysis of DCE T1-w, and DW automatically registered images. Proc. SPIE 8670, Medical Imaging 2013: Computer-Aided Diagnosis, 86703Google Scholar
- 11.Matulewicz L et al (2013) Anatomic segmentation improves prostate cancer detection with artificial neural networks analysis of H magnetic resonance spectroscopic imaging. J Magn Reson Imaging 1421:1414–1421Google Scholar
- 12.Tiwari P, Kurhanewicz J, Rosen M, Madabhushi A (2010) Semi supervised multi-kernel (SeSMiK) graph embedding: identifying aggressive prostate cancer via magnetic resonance imaging and spectroscopy. Int Conf Med Image Comput Comput Interv 13:666–67Google Scholar
- 16.Tofts PS, Brix G, Buckley DL, Evelhoch JL, Henderson E, Knopp MV, H. B. W. Larsson, Lee T-Y, Mayr NA, G. J. M. Parker, Port RE, Taylor J, Weiskoff R (1999) Estimating kinetic parameters from dynamic contrast-enhanced T1-weighted MRI of a diffusible tracer: Standardized quantities and symbols. J Magn Reson Imaging 10:223–232CrossRefPubMedGoogle Scholar
- 17.Wang S, Burtt K, Turkbey B, Choyke P, Summers RM (2014) Computer aided-diagnosis of prostate cancer on multiparametric MRI: a technical review of current research. BioMed Res IntGoogle Scholar
- 19.Artan Y et al (2009) Prostate cancer segmentation with multispectral MRI using cost-sensitive conditional random fields. In: 2009 IEEE international symposium on biomedical imaging: from Nano to Macro, Boston, MA, pp 278–281Google Scholar
- 20.Liu P et al (2013) A prostate cancer computer-aided diagnosis system using multimodal magnetic resonance imaging and targeted biopsy labels. In: Proceedings of SPIE—the international society for optical engineering, vol 8670Google Scholar
- 24.Litjens GJS, Barentsz JO, Karssemeijer N & Huisman HJ (2012) Automated computer-aided detection of prostate cancer in MR images: from a whole-organ to a zone-based approach. In: Progress in biomedical optics and imaging—proceedings of SPIE, vol 8315Google Scholar
- 26.Ampeliotis D, Antonakoudi A, Berberidis K, Psarakis EZ (2007) Computer aided detection of prostate cancer using fused information from dynamic contrast enhanced and morphological magnetic resonance images. In: IEEE international conference on signal processing and communications, 2007. ICSPC 2007, Dubai, pp 888–891Google Scholar
- 27.Karatzoglou A, Meyer D, Hornik K (2006) Support vector machines. R J Stat Softw 15:28Google Scholar
- 28.Pedregosa F et al (2011) Scikit-learn: Machine learning in Python. J Mach Learn Res 12:2825–2830Google Scholar
- 31.Tofts PS T1-weighted DCE imaging concepts: modelling, acquisition and analysis. Signal 500(450):400Google Scholar
- 32.Shen Y, Snyder C, Goerner F, Moritz R, Runge V (2013) Accurate T1 relaxivities (r1) of gadolinium-based magnetic resonance contrast agents (GBCAs) in human whole blood at 1.5 T and 3 T. Radiological Society of North America 2013 Scientific Assembly and Annual MeetingGoogle Scholar
- 34.Whitcher B, Schmid V, Thornton A (2011) oro. nifti: Rigorous-NIfTI input/output. R package version 0. 2:6Google Scholar
- 35.R Core Team (2014) R: a language and environment for statistical computing. R Foundation for Statistical Computing, ViennaGoogle Scholar
- 36.Boser BE, Guyon IM, Vapnik VN (1992) A training algorithm for optimal margin classifiers. In: Proceedings of the 5th annual ACM workshop on computing learning theory, pp 144–152Google Scholar
- 38.Hofmann T, Schölkopf B, Smola AJ (2006) A review of kernel methods in machine learning. Mac-Planck-Institut für biologische, Kybernetik, Tech. Rep, p 156Google Scholar
- 40.Hsu CW, Chang CC, Lin CJ (2003) A practical guide to support vector classificationGoogle Scholar
- 42.Liaw A, Wiener M (2002) Classification and regression by random Forest. R news 2:18–22Google Scholar
- 54.Rossi F et al (2015) A 3D voxel neighborhood classification approach within a multiparametric MRI Classifier for prostate cancer detection, pp 231–239Google Scholar