Cascaded convolutional neural networks for spine chordoma tumor segmentation from MRI

Chordoma is a rare type of tumor that usually appears in the bone near the spinal cord and skull base. Due to their location in the skull base and diverse appearance in size and shape, automatic segmentation of chordoma tumors from magnetic resonance images (MRI) is a challenging task. In addition, similar MR intensity distributions of different anatomical regions, specifically sinuses, make the segmentation task from MRI more challenging. In comparison, most of the state-of-the-art lesion segmentation methods are designed to segment pathologies inside the brain. In this work, we propose an automatic chordoma segmentation framework using two cascaded 3D convolutional neural networks (CNN) via an auto-context model. While the first network learns to detect all potential tumor voxels, the second network fine-tunes the classifier to distinguish true tumor voxels from the false positives detected by the first network. The proposed method is evaluated using multi-contrast MR images of 22 longitudinal scans from 8 patients. Preliminary results showed a linear correlation of 0.71 between the detected and manually outlined tumor volumes, compared to 0.40 for a random forest (RF) based method. Furthermore, the response of tumor growth over time, i.e. increasing, decreasing, or stable, is evaluated according to the response evaluation criteria in solid tumors with an outcome of 0.26 kappa coefficient, compared to 0.13 for the RF based method.

[1]  Robert Ford,et al.  Individual patient data analysis to assess modifications to the RECIST criteria. , 2009, European journal of cancer.

[2]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[3]  Marti A. Hearst Trends & Controversies: Support Vector Machines , 1998, IEEE Intell. Syst..

[4]  Alex Rovira,et al.  Improving automated multiple sclerosis lesion segmentation with a cascaded 3D convolutional neural network approach , 2017, NeuroImage.

[5]  Brian B. Avants,et al.  N4ITK: Improved N3 Bias Correction , 2010, IEEE Transactions on Medical Imaging.

[6]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.

[7]  Arno Klein,et al.  A reproducible evaluation of ANTs similarity metric performance in brain image registration , 2011, NeuroImage.

[8]  Vladimir Fonov,et al.  Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: The CADDementia challenge , 2015, NeuroImage.

[9]  Jyrki Lötjönen,et al.  Robust whole-brain segmentation: Application to traumatic brain injury , 2015, Medical Image Anal..

[10]  Zhuowen Tu,et al.  Auto-Context and Its Application to High-Level Vision Tasks and 3D Brain Image Segmentation , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Snehashis Roy,et al.  Subject-Specific Sparse Dictionary Learning for Atlas-Based Brain MRI Segmentation , 2015, IEEE Journal of Biomedical and Health Informatics.

[12]  Bennett A Landman,et al.  Out-of-atlas likelihood estimation using multi-atlas segmentation. , 2013, Medical physics.

[13]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[14]  Snehashis Roy,et al.  Longitudinal multiple sclerosis lesion segmentation: Resource and challenge , 2017, NeuroImage.

[15]  et al.,et al.  ISLES 2015 ‐ A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI , 2017, Medical Image Anal..