Optimising a Simple Fully Convolutional Network for Accurate Brain Age Prediction in the PAC 2019 Challenge
Bibliographic Data
| ID | 15529848 |
|---|---|
| Authors | Weikang Gong (0000-0002-5387-4428, Wellcome Centre for Integrative Neuroimaging), Christian F Beckmann (0000-0002-3373-3193, Radboud University Nijmegen), Andrea Vedaldi (0000-0003-1374-2858, University of Oxford), Stephen M Smith (0000-0001-8166-069X, Wellcome Centre for Integrative Neuroimaging), Han Peng (0000-0002-1880-1186, Wellcome Centre for Integrative Neuroimaging, corresponding author) |
| Year | 2021 |
| Volume | 12 |
| Pages | 627996-627996 |
| Publication date | 2021-05-10 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Frontiers in Psychiatry (JOURNAL) |
| Journal identifiers | ISSN: 1664-0640 • E-ISSN: 1664-0640 |
| Publisher | Frontiers Media (PUBLISHER • CH) |
| DOI | 10.3389/fpsyt.2021.627996 |
| PMID | 34040552 |
| OpenAlex | W3161371747 |
| Language | EN |
| References cited | 32 |
Brain age prediction from brain MRI scans not only helps improve brain ageing modelling generally, but also provides benchmarks for predictive analysis methods. Brain-age delta, which is the difference between a subject's predicted age and true age, has become a meaningful biomarker for the health of the brain. Here, we report the details of our brain age prediction models and results in the Predictive Analysis Challenge 2019. The aim of the challenge was to use T1-weighted brain MRIs to predict a subject's age in multicentre datasets. We apply a lightweight deep convolutional neural network architecture, Simple Fully Convolutional Neural Network (SFCN), and combined several techniques including data augmentation, transfer learning, model ensemble, and bias correction for brain age prediction. The model achieved first place in both of the two objectives in the PAC 2019 brain age prediction challenge: Mean absolute error (MAE) = 2.90 years without bias removal (Second Place = 3.09 yrs; Third Place = 3.33 yrs), and MAE = 2.95 years with bias removal, leading by a large margin (Second Place = 3.80 yrs; Third Place = 3.92 yrs)
Convolutional neural network · Machine learning · Margin (machine learning · Computer Science · Domain Adaptation and Few-Shot Learning · Functional Brain Connectivity Studies · Neonatal and fetal brain pathology · Artificial Intelligence
ImageNet classification with deep convolutional neural networks
To Explain or to Predict?
ImageNet Large Scale Visual Recognition Challenge
Deep learning
Behavioral and Psychological Effects of Coronavirus Disease-19 Quarantine in Patients With Dementia
Classifying Autism Spectrum Disorder Using the Temporal Statistics of Resting-State Functional MRI Data With 3D Convolutional Neural Networks
| Highly cited | No |
|---|