Predicting Alzheimer’s Disease Progression

with Deep Generative Models

from MRI Data

presented by:

[Hamid Salehian]

[Prof. Babak Majidi]

Supervisor:

Academic Year

1396 - 1397

Introduction

AlZheimer's Disease: What

• Alzheimer’s is an irreversible, progressive brain disease that slowly destroys memory and disorder cognitive function.
• Clinical studies can indicate disease first symptoms of cognitive impairment while the disease is already in an dvanced stage
• Sixth leading cause of death in the United States.
• One out of 85 persons will have                                              Alzheimers disease (AD) by 2050

AlZheimer's Disease: Why

• Plaques. These clumps of a protein called beta-amyloid may damage and destroy brain cells in several ways, including interfering with inter-cell communication.

• Tangles.  Twisted fibers of a protein called tau. disintegrate the transport system so nutrients and other essential supplies can no longer move through the cells, which eventually die.

AlZheimer's Disease: Causes

Dying cells cause brain shrinkage

AlZheimer’s Diagnostic

• Review medical history
• Mini Mental Status Exam (MMSE)
• Physical Exam
• Neurological Exam
• Brain Image: Structural(MRI,CT), Functional(fMRI), PET (Positron Emission Tomography)
• Cerebrospinal Fluid (CSF)

Allows for measurement of the 3-dimensional (3D) volume of brain structures, especially the size of the hippocampus and ventricles enlargement

MRI: Magnetic resonance imaging

Previous Works

 Diagnosis Methods

Automatic methods:

volumetric and shape features together with PCA and SVM were used to
classify MRI images as having the disease or not (Lee, et al., 2009).

Effect of PCA on SVM classfier and k-means Clustering for Alzheimer's Disease Diagnosis

Deep Learning: using 3D Convolutional Neural Networks for classification. (Korelev 2017).

CAE: Convolutional Autoencoder

SAE: Sparse Autoencoder

CNN: Convolutional Neural Networks

MSLF: Matrix-Similarity based Loss Function

MTFS: Multi-task feature selection

MFE: Multiview Feature Extraction

PCA: Principal Component Analysis

 Diagnosis Methods

CVAE: Convolutional Variational Autoencoders

CAAE:Conditional Adversarial Autoencoders

SAE: Sparse Autoencoder

CNN: Convolutional Neural Networks

PCA: Principal Component Analysis

SVM: Support Vector Machine

MDL: Manifold Deep Learning

 EarlY Diagnosis Methods

GOALS

Solutions

• Image reconstruction by domain-transform
• Deep Generative Model to improve MRI images
• Predict Disease progression by reconsrtructed Images

Zhu et al: Image reconstruction by domain-transform
manifold learning (2017) (AUTOMAP)

Khajehnejad et al: Alzheimer’s Disease Early Diagnosis Using
Manifold-Based Semi-Supervised Learning (2017)

Datasets

 Online Datasets

HARP:  European Alzheimer’s Disease Consortium to gether with the Alzheimer’s Disease Neuroimaging Initiative (ADNI). It was provided as HARmonized Protocol (HARP) for manual hippocampal segmentation from MRI  and it consists of 131 volumes.

ADNI: Alzheimer’s Disease Neuroimaging Initiative

OASIS: the Open Access Series of Imaging Studies (OASIS) was created by the Washington University Alzheimer’s Disease Research Center, It consists of 233 volumes.

References

[1] B. Birur, N. V. Kraguljac, R. C. Shelton, and A. C. Lahti. Brain structure, function, and neurochemistry in schizophrenia and bipolardisorder a systematic review of the magnetic resonance neuroimaging literature. NPJ Schizophrenia, 3, 15, 2017.
[2] Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, and L. D. Jackel. Handwritten digit recognition with a back-propagation network.NIPS, 1989.
[3] A. Krizhevsky, I. Sutskever, and G. E. Hinton. Imagenet classificationwith deep convolutional neural networks. Advances in neural information processing systems (NIPS), 1097-1105, 2012.
[4] P. Bojanowski, A. Joulin, D. Lopez-Paz, and A. Szlam. Optimizing theLatent Space of Generative Networks. arXiv:1707.05776v1 [stat.ML], 2017.
[5] A. Radford, L. Metz, and S. Chintala. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. arXiv:1511.06434v2 [cs.LG], 2015.
[6] T. D. Kulkarni, W.F . Whitney, P. Kohli, and J. B. Tenenbaum. Deep Convolutional Inverse Graphics Network. Advances in Neural Informa- tion Processing Systems 28 (NIPS ), 2015.
[7] A. Achlioptas, O. Diamanti, I. Mitliagkas, and L. Guibas. Representation Learning and Adversarial Generation of 3D Point Clouds. arXiv:1707.02392v1 [cs.CV], 2017.
[8] Z. Zhang, Y. Song, and H. Qi. Age progression/regression by con-ditional adversarial autoencoder. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.

References

[9] M. R. Arbabshirani, S. Plis, J. Sui, and V. D. Calhoun. Single subject prediction of brain disorders in neuroimaging: Promises and pitfalls. Neuroimage; 145(Pt B):137-165, 2016.
[10] M. Liu, E. Zhang, D. Adeli-Mosabbeb, and D. Shen. Inherent Structure Based Multi-view Learning with Multi-template Feature Representation for Alzheimer’s Disease Diagnosis. IEEE Trans BiomedicalEngineering; 63(7): 14731482, 2016.
[11] N. Amoroso, M. La Rocca, s: Bruno, T. Maggipinto, A. Monaco, R. Bellotti, and S. Tangaro. Brain structural connectivity atrophy in Alzheimer’s disease.
[12] R. Brosch, T. andTam and Initiative for the Alzheimer’s Disease Neuroimaging. Manifold learning of brain MRIs by deep learning. MICCAI16 (Pt 2):633-40, 2013.
[13] E. Hosseini-Asl, R. Keynton, and A. El-Baz. Alzheimer’s disease diagnostics by adaptation of 3D convolutional network. IEEE International Conference on Image Processing (ICIP), 2016.
[14] L. G. Apostolova, A. E. Green, S. Babakchanian, K. S. Hwang, Y.-Y. Chou, A. W. Toga, and P. M. Thompson. Hippocampal atrophy and ventricular enlargement in normal aging, mild cognitive impairment and Alzheimer’s disease. Alzheimer’s Disease and Associated Disorders, 26(1), 17 - 27, 2012.
[15] A. G. Roy et al. Error Corrective Boosting for Learning Fully Con- volutional Networks with Limited Data. Proceedings Medical Image Computing and Computer-Assisted Intervention - MICCAI 2017 - 20th International Conference, 2017.
[16] 2017 Alzheimer’s Disease Facts and Figures. Alzheimer’s Disease As- sociation, 2017.
[17] L. G. Apostolova. Alzheimer’s Disease. Continuum: Lifelong Learning in Neurology 22. 2 Dementia, 419-434, 2016.
[18] D. P. Kingma and M. Welling. Auto-Encoding Variational Bayes. Pro- ceedings of the 2nd International Conference on Learning Representations (ICLR), 2014.

Question!?

Thank You For Your Attention

powered by: slides.com

see more: https://slides.com/hsarena