U-net model for brain extraction: Trained on humans for transfer to non-human primates
Xindi Wang,
Xin-Hui Li,
Jae Wook Cho,
Brian E. Russ,
Nanditha Rajamani,
Alisa Omelchenko,
Lei Ai,
Annachiara Korchmaros,
Stephen Sawiak,
R. Austin Benn,
Pamela Garcia-Saldivar,
Zheng Wang,
Ned H. Kalin,
Charles E. Schroeder,
R. Cameron Craddock,
Andrew S. Fox,
Alan C. Evans,
Adam Messinger,
Michael P. Milham,
Ting Xu
Affiliations
Xindi Wang
Montreal Neurological Institute, McGill University, Montreal, Québec, Canada; Corresponding authors.
Xin-Hui Li
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA
Jae Wook Cho
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA
Brian E. Russ
Nathan Kline Institute, 140 Old Orangeburg Rd, Orangeburg, NY, USA; Department of Neuroscience, Icahn School of Medicine at Mount Sinai, New York City, NY, USA; Department of Psychiatry, New York University School of Medicine, New York City, NY, USA
Nanditha Rajamani
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA
Alisa Omelchenko
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA
Lei Ai
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA
Annachiara Korchmaros
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA
Stephen Sawiak
Translational Neuroimaging Laboratory, Department of Physiology, Development and Neuroscience University of Cambridge, Cambridge CB2 3EG, UK
R. Austin Benn
Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain
Pamela Garcia-Saldivar
Instituto de Neurobiología, Universidad Nacional Autónoma de México Campus Juriquilla, Querétaro, México
Zheng Wang
Institute of Neuroscience, CAS Center for Excellence in Brain Science and Intelligence Technology, State Key Laboratory of Neuroscience, CAS Key Laboratory of Primate Neurobiology, Chinese Academy of Science, Shanghai, China; University of Chinese Academy of Science, China
Ned H. Kalin
Department of Psychiatry, University of Wisconsin School of Medicine and Public Health, 6001 Research Park Blvd, Madison, WI 53719, USA
Charles E. Schroeder
Nathan Kline Institute, 140 Old Orangeburg Rd, Orangeburg, NY, USA; Departments of Psychiatry and Neurology, Columbia University College of Physicians and Surgeons, New York, NY 10032, USA
R. Cameron Craddock
Department of Diagnostic Medicine, The University of Texas at Austin Dell Medical School, USA
Andrew S. Fox
Department of Psychology, and the California National Primate Research Center, University of California, Davis, One Shields Ave., Davis, CA 95616, USA
Alan C. Evans
Montreal Neurological Institute, McGill University, Montreal, Québec, Canada
Adam Messinger
Laboratory of Brain and Cognition, National Institute of Mental Health, Bethesda, USA
Michael P. Milham
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA; Nathan Kline Institute, 140 Old Orangeburg Rd, Orangeburg, NY, USA
Ting Xu
The Child Mind Institute, 101 East 56th Street, New York, NY 10022, USA; Corresponding authors.
Brain extraction (a.k.a. skull stripping) is a fundamental step in the neuroimaging pipeline as it can affect the accuracy of downstream preprocess such as image registration, tissue classification, etc. Most brain extraction tools have been designed for and applied to human data and are often challenged by non-human primates (NHP) data. Amongst recent attempts to improve performance on NHP data, deep learning models appear to outperform the traditional tools. However, given the minimal sample size of most NHP studies and notable variations in data quality, the deep learning models are very rarely applied to multi-site samples in NHP imaging. To overcome this challenge, we used a transfer-learning framework that leverages a large human imaging dataset to pretrain a convolutional neural network (i.e. U-Net Model), and then transferred this to NHP data using a small NHP training sample. The resulting transfer-learning model converged faster and achieved more accurate performance than a similar U-Net Model trained exclusively on NHP samples. We improved the generalizability of the model by upgrading the transfer-learned model using additional training datasets from multiple research sites in the Primate Data-Exchange (PRIME-DE) consortium. Our final model outperformed brain extraction routines from popular MRI packages (AFNI, FSL, and FreeSurfer) across a heterogeneous sample from multiple sites in the PRIME-DE with less computational cost (20 s~10 min). We also demonstrated the transfer-learning process enables the macaque model to be updated for use with scans from chimpanzees, marmosets, and other mammals (e.g. pig). Our model, code, and the skull-stripped mask repository of 136 macaque monkeys are publicly available for unrestricted use by the neuroimaging community at https://github.com/HumanBrainED/NHP-BrainExtraction.
