Fader networks for domain adaptation on fMRI: ABIDE-II study

ABIDE is the largest open-source autism spectrum disorder database with both fMRI data and full phenotype description. These data were extensively studied based on functional connectivity analysis as well as with deep learning on raw data, with top models accuracy close to 75% for separate scanning sites. Yet there is still a problem of models transferability between different scanning sites within ABIDE. In the current paper, we for the first time perform domain adaptation for brain pathology classification problem on raw neuroimaging data. We use 3D convolutional autoencoders to build the domain irrelevant latent space image representation and demonstrate this method to outperform existing approaches on ABIDE data.

[1]  Margot J. Taylor,et al.  Altered structural brain asymmetry in autism spectrum disorder in a study of 54 datasets , 2019, Nature Communications.

[2]  Evgeny Burnaev,et al.  Voxelwise 3D Convolutional and Recurrent Neural Networks for Epilepsy and Depression Diagnostics from Structural and Functional MRI Data , 2018, 2018 IEEE International Conference on Data Mining Workshops (ICDMW).

[3]  Dimitris Samaras,et al.  Deriving reproducible biomarkers from multi-site resting-state data: An Autism-based example , 2016, NeuroImage.

[4]  Amirali Kazeminejad,et al.  Topological Properties of Resting-State fMRI Functional Networks Improve Machine Learning-Based Autism Classification , 2019, Front. Neurosci..

[5]  Shantenu Jha,et al.  Learning Neural Markers of Schizophrenia Disorder Using Recurrent Neural Networks , 2017, ArXiv.

[6]  Dinggang Shen,et al.  Multiple-Network Classification of Childhood Autism Using Functional Connectivity Dynamics , 2014, MICCAI.

[7]  Khundrakpam Budhachandra,et al.  The Neuro Bureau Preprocessing Initiative: open sharing of preprocessed neuroimaging data and derivatives , 2013 .

[8]  Guido A. van Wingen,et al.  A Hybrid 3DCNN and 3DC-LSTM Based Model for 4D Spatio-Temporal fMRI Data: An ABIDE Autism Classification Study , 2019, OR/MLCN@MICCAI.

[9]  Takeo Watanabe,et al.  A small number of abnormal brain connections predicts adult autism spectrum disorder , 2016, Nature Communications.

[10]  Victor S. Lempitsky,et al.  Unsupervised Domain Adaptation by Backpropagation , 2014, ICML.

[11]  Mert R. Sabuncu,et al.  Ensemble learning with 3D convolutional neural networks for functional connectome-based prediction , 2019, NeuroImage.

[12]  Manhua Liu,et al.  Classification of Alzheimer's Disease by Cascaded Convolutional Neural Networks Using PET Images , 2017, MLMI@MICCAI.

[13]  Nicha C. Dvornek,et al.  Multi-site fMRI Analysis Using Privacy-preserving Federated Learning and Domain Adaptation: ABIDE Results , 2020, Medical Image Anal..

[14]  Daniel P. Kennedy,et al.  Enhancing studies of the connectome in autism using the autism brain imaging data exchange II , 2017, Scientific Data.

[15]  Guillaume Lample,et al.  Fader Networks: Manipulating Images by Sliding Attributes , 2017, NIPS.