A deep learning based approach to classification of CT brain images

This study explores the applicability of the state of the art of deep learning convolutional neural network (CNN) to the classification of CT brain images, aiming at bring images into clinical applications. Towards this end, three categories are clustered, which contains subjects' data with either Alzheimer's disease (AD) or lesion (e.g. tumour) or normal ageing. Specifically, due to the characteristics of CT brain images with larger thickness along depth (z) direction (~5mm), both 2D and 3D CNN are employed in this research. The fusion is therefore conducted based on both 2D CT images along axial direction and 3D segmented blocks with the accuracy rates are 88.8%, 76.7% and 95% for classes of AD, lesion and normal respectively, leading to an average of 86.8%.

[1]  Y. LeCun,et al.  Learning methods for generic object recognition with invariance to pose and lighting , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[2]  Kunihiko Fukushima,et al.  Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position , 1980, Biological Cybernetics.

[3]  K. Engedal,et al.  Accuracy of CT scan measurements of the medial temporal lobe in routine dementia diagnostics , 2003, International journal of geriatric psychiatry.

[4]  Yi Zhang,et al.  Usefulness of computed tomography linear measurements in diagnosing alzheimer's disease , 2008, Acta radiologica.

[5]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[6]  Andrea Vedaldi,et al.  MatConvNet: Convolutional Neural Networks for MATLAB , 2014, ACM Multimedia.

[7]  Ming Yang,et al.  3D Convolutional Neural Networks for Human Action Recognition , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Marc'Aurelio Ranzato,et al.  Unsupervised Learning of Invariant Feature Hierarchies with Applications to Object Recognition , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.