Convolution neural network has huge potential for accurate disease prediction with neuroimaging data, but the popular ImageNet models are often not transferable to medical data due to the training-dataset size and computing memory requirements. To address this, we propose a deep convolutional neural network model, Simple Fully Convolutional Network (SFCN), benchmarked with brain age prediction using 3D T1-weighted structural MRI data. Compared with other popular deep network architectures, SFCN has fewer parameters, so is more compatible with small dataset size and 3D volume data. We compared our SFCN approach with several widely-used machine learning models. It achieved state-of-the-art performance in UK Biobank brain imaging data (N = 14,503), with mean absolute error (MAE) = 2.14y in brain age prediction and 99.5% in sex classification. SFCN also won the 2019 Predictive Analysis Challenge for brain age prediction, involving 79 competing teams (N = 2,638, MAE = 2.90y).
In this talk, Dr. Peng will introduce the design and the optimisation of SFCN, and brain ageing pattern discovery with interpreting internal activations of convnet predictive models using the large UK Biobank brain imaging dataset.
Reference
Code and pretrained weights
https://github.com/ha-ha-ha-han/UKBiobank_deep_pretrain