DCE-MRI Data Analysis for Cancer Area Classification

 

 Buy Article Permissions and Reprints

Summary

Objectives: The paper aims at improving the support of medical researchers in the context of in-vivo cancer imaging. Morphological and functional parameters obtained by dynamic contrast-enhanced MRI (DCE-MRI) techniques are analyzed, which aim at investigating the development of tumor microvessels. The main contribution consists in proposing a machine learning methodology to segment automatically these MRI data, by isolating tumor areas with different meaning, in a histological sense.

Methods: The proposed approach is based on a three-step procedure: i) robust feature extraction from raw time-intensity curves, ii) voxel segmentation, and iii) voxel classification based on a learning-by-example approach. In the first step, few robust features that compactly represent the response of the tissue to the DCE-MRI analysis are computed. The second step provides a segmentation based on the mean shift (MS) paradigm, which has recently shown to be robust and useful for different and heterogeneous clustering tasks. Finally, in the third step, a support vector machine (SVM) is trained to classify voxels according to the labels obtained by the clustering phase (i.e., each class corresponds to a cluster). Indeed, the SVM is able to classify new unseen subjects with the same kind of tumor.

Results: Experiments on different subjects affected by the same kind of tumor evidence that the extracted regions by both the MS clustering and the SVM classifier exhibit a precise medical meaning, as carefully validated by the medical researchers. Moreover, our approach is more stable and robust than methods based on quantification of DCE-MRI data by means of pharmacokinetic models.

Conclusions: The proposed method allows to analyze the DCE-MRI data more precisely and faster than previous automated or manual approaches.

• References

• 1 Lau PY, Ozawa S, Voon FCT. Using Block-based Multiparameter Representation to Detect Tumor Features on T2-weighted Brain MRI Images. Methods Inf Med 2007; 46: 236-241.
  Article in Thieme ConnectPubMedGoogle Scholar
• 2 Duda RO, Hart PE, Stork DG. Pattern Classification. John Wiley and Sons, second edition; 2001
  Google Scholar
• 3 Castellani U, Cristani M, Combi C, Murino V, Sbarbati A, Marzola P. Visual MRI: Merging information visualization and non-parametric clustering techniques for MRI dataset analysis. Artificial Intelligence in Medicine 2008; 44 (03) 171-282.
  CrossrefPubMedGoogle Scholar
• 4 Buckley DL. Uncertainty in the analysis of tracer kinetics using dynamic contrast enhanced T1-weighted MRI. Magnetic Resonance Med 2002; 47: 601-606.
  CrossrefPubMedGoogle Scholar
• 5 Harrer JU, Parker GJ, Haroon HA, Buckley DL, Embelton K, Roberts C. et al. Comparative study of methods for determining vascular permeability and blood volume in human gliomas. Magnetic Resonance Imaging 2004; 20: 748-757.
  CrossrefPubMedGoogle Scholar
• 6 Lavinia C, De Jongea MC, Van de Sandeb MGH, Takb PP, Nederveena AJ, Maas M. Pixel-by-pixel analysis of DCE MRI curve patterns and an illustration of its application to the imaging of the musculoskeletal system. Magnetic Resonance Imaging 2007; 25: 604-612.
  CrossrefPubMedGoogle Scholar
• 7 Tofts PS, Briks G, Buckley DL, Evelhoch JL, Henderson E, Knopp MV. et al. Estimating kinetic parameters from dynamic contrast enhanced T1-w MRI of a diffusible tracer: standardized quantities and symbols. Magnetic Resonance Imaging 1999; 10: 223-232.
  CrossrefPubMedGoogle Scholar
• 8 Comaniciu D, Meer P. Mean shift: A robust approach toward feature space analysis. IEEE Trans Pattern Anal Mach Intell 2002; 24: 603-619.
  CrossrefPubMedGoogle Scholar
• 9 Burges C. A tutorial on support vector machine for pattern recognition. Data Mining and Knowledge Discovery 1998; 2: 121-167.
  CrossrefPubMedGoogle Scholar
• 10 Stoutjesdijk MJ, Veltman J, Huisman MD, Karssemeijer N, Barents J. et al. Automatic analysis of contrast enhancement in breast MRI lesions using mean shift clustering for ROI selection. Journal of Magnetic Resonance Imaging 2007; 26: 606-614.
  CrossrefPubMedGoogle Scholar
• 11 Twellmann T, Saalbach A, Muller C, Nattkemper TW, Wismuller A. Detection of suspicious lesions in dynamic contrast-enhanced MRI data. Engineering in Medicine and Biology Society. 2004 pp 454-457.
  PubMedGoogle Scholar
• 12 Leistritz L, Hesse W, Wustenberg T, Fitzek C, Reichenbach JR, Witte H. Time-variant analysis of fast-fMRI and dynamic contrast agent MRI sequences as examples of 4-dimensional image analysis. Methods Inf Med 2006; 45: 643-650.
  Article in Thieme ConnectPubMedGoogle Scholar
• 13 Marzola P, Ramponi S, Nicolato E, Lovati E, Sandri M, Calderan L, Crescimanno C, Merigo F, Sbarbati A, Grotti A, Vultaggio S, Cavagna F, Lo Russo V, Osculati F. Effect of tamoxifen in an experimental model of breast tumor studied by dynamic contrast-enhanced magnetic resonance imaging and different contrast agents. Investigative radiology 2005; 40: 421-429.
  CrossrefPubMedGoogle Scholar
• 14 Marzola P, Degrassi A, Calderan L, Farace P, Crescimanno C, Nicolato E, Giusti A, Pesenti E, Terron A, Sbarbati A, Abrams T, Murray L, Osculati F. In vivo assessment of antiangiogenic activity of su6668 in an experimental colon carcinoma model. Clin Cancer Res 2004; 2: 739-50.
  PubMedGoogle Scholar
• 15 Chang CC, Lin CJ. LIBSVM: a library for support vector machines. 2001
  PubMedGoogle Scholar