DeepDefacer: Automatic Removal of Facial Features via U-Net Image Segmentation

  title={DeepDefacer: Automatic Removal of Facial Features via U-Net Image Segmentation},
  author={Anish Khazane and Julien Hoachuck and Krzysztof J. Gorgolewski and Russell A. Poldrack},
Recent advancements in the field of magnetic resonance imaging (MRI) have enabled large-scale collaboration among clinicians and researchers for neuroimaging tasks. However, researchers are often forced to use outdated and slow software to anonymize MRI images for pub-lication. These programs specifically perform expensive mathematical operations over 3D images that rapidly slows down anonymization speed as an image’s volume increases in size. In this paper, we introduce DeepDefacer, an… 
1 Citations

Figures and Tables from this paper

Segmentation stability of human head and neck cancer medical images for radiotherapy applications under de-identification conditions: Benchmarking data sharing and artificial intelligence use-cases

The choice of defacing algorithm has a significant impact on HNC OAR auto-segmentation model development and deployment and incentivizes the further development of HNC radiotherapy-specific defacing methods.



U-Net: Convolutional Networks for Biomedical Image Segmentation

It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.

A technique for the deidentification of structural brain MR images

Results suggested that the automatic defacing algorithm is robust, efficiently removes nonbrain tissue, and does not unduly influence the outcome of the processing methods utilized; in some cases, skull‐stripping was improved.

Performance evaluation of 2D and 3D deep learning approaches for automatic segmentation of multiple organs on CT images

All the deep learning approaches showed a better accuracy and robustness than the conventional segmentation method that used probabilistic atlas and graph-cut methods for solving multiple organs segmentation problem on 3D CT images.

Preserving Privacy in Structural Neuroimages

The Quickshear Defacing method uses a convex hull to identify a plane that divides the volume into two parts, one containing facial features and another the brain volume, and removes the voxels on the facial features side and can provide reductions in running time.

3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation

The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts and performs on-the-fly elastic deformations for efficient data augmentation during training.

A hybrid approach to the Skull Stripping problem in MRI

A Non-Technical Survey on Deep Convolutional Neural Network Architectures

This survey gives a comprehensive introduction to this topic (object recognition with deep convolutional neural networks), with a strong focus on the evolution of network architectures.

Fully convolutional networks for semantic segmentation

The key insight is to build “fully convolutional” networks that take input of arbitrary size and produce correspondingly-sized output with efficient inference and learning.

HIPAA administrative simplification: standard unique health identifier for health care providers. Final rule.

  • Friday
  • Medicine, Political Science
    Federal register
  • 2004
The use of the NPI will improve the Medicare and Medicaid programs, other Federal health programs and private health programs, and the effectiveness and efficiency of the health care industry in general, by simplifying the administration of thehealth care system and enabling the efficient electronic transmission of certain health information.

The National Institute of Biomedical Imaging and Bioengineering.

  • S. Chien
  • Biology, Engineering
    Annual review of biomedical engineering
  • 2004
Tissue engineering evolved from the field of biomaterials development and refers to combining scaffolds, cells, and biologically active molecules into functional tissues that restore, maintain, or improve damaged tissue or a whole organ.