kevin-keraudren / fetus-detector

Brain and body detectors for fetal MRI developed during my PhD

Automated Organ Localisation in Fetal Magnetic Resonance Imaging

Code developed throughout my PhD in order to automatically find the organs of a fetus in MR images:

• brain-detector [1,2]
• body-detector [3]
• commonlib contains code used in both projects, in particular fetal biometric measurements in commonlib/fetal_anatomy.py
• external contains code that is not used in [1,2,3] but is relevant to automated detection in fetal MRI, such as the RIF features used in [4]

This code is based on python-irtk, a Python interface to IRTK, which can be installed through conda:

conda install -c kevin-keraudren python-irtk

It uses OpenCV version 2 (note that SIFT features are not included by default in OpenCV 3 binaries).

Compiling the code

cd body-detector && make

Running the demo

cd body-detector && ./demo.sh

Below is a screenshot of the detection results (detection_results/stack-1/prediction_2/final_seg.nii.gz):

A demo focusing on the brain detection is available in the repository example-motion-correction.

References

[1] Keraudren, K., Kyriakopoulou, V., Rutherford, M., Hajnal, J.V., & Rueckert, D.: Localisation of the Brain in Fetal MRI Using Bundled SIFT Features. MICCAI 2013.
PDF video poster

[2] Keraudren, K., Kuklisova-Murgasova, M., Kyriakopoulou, V., Malamateniou, C., Rutherford, M.A., Kainz, B., Hajnal, J.V., & Rueckert, D.: Automated Fetal Brain Segmentation from 2D MRI Slices for Motion Correction. NeuroImage, 2014.
PDF demo slides

[3] Keraudren, K., Kainz, B., Oktay, O., Kyriakopoulou, V., Rutherford, M., Hajnal, J. V., & Rueckert, D.: Automated Localization of Fetal Organs in MRI Using Random Forests with Steerable Features. MICCAI 2015.
PDF video slides

[4] Kainz, B., Keraudren, K., Kyriakopoulou, V., Rutherford, M., Hajnal, J., & Rueckert, D.: "Fast Fully Automatic Brain Detection in Fetal MRI Using Dense Rotation Invariant Image Descriptors". ISBI 2014.
PDF video