Where is my baby? A fast fetal head auto-alignment in 3D-ultrasound

Ultrasonography is the main modality for prenatal screening examination of the fetal central nervous system. Due to the variability of the position of the fetal brain relatively to the probe, getting through structures of interest requires both time and a high level of expertise. The method presented in this paper aims at helping the clinician navigate through the brain by automatically aligning the head in near real time (<; 1 s) in a 3D ultrasound volume. The alignment is obtained by defining a frame of reference based on the skull, the mid-sagittal plan and the orbits of the eyes; their signals remain strong and stable across acquisitions. They are detected by combining state-of-the-art techniques (random forests and template deformation). Our method has proven fast and accurate on a dataset of 78 volumes (19-24 gestational weeks): maximal alignment errors' medians range from 5.1 to 5.8mm for the transcerebellar, transventricular and transthalamic planes.

[1]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[2]  J. Alison Noble,et al.  Ultrasound image segmentation: a survey , 2006, IEEE Transactions on Medical Imaging.

[3]  Zygmunt L. Szpak,et al.  Guaranteed Ellipse Fitting with the Sampson Distance , 2012, ECCV.

[4]  Raghu Machiraju,et al.  Anisotropic plate diffusion filtering for detection of cell membranes in 3D microscopy images , 2010, 2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[5]  Isabelle Bloch,et al.  Brain symmetry plane computation in MR images using inertia axes and optimization , 2002, Object recognition supported by user interaction for service robots.

[6]  Dorin Comaniciu,et al.  Automatic fetal face detection from ultrasound volumes via learning 3D and 2D information , 2009, CVPR.

[7]  Nassir Navab,et al.  Automatic CT-ultrasound registration for diagnostic imaging and image-guided intervention , 2008, Medical Image Anal..

[8]  J. Shotton,et al.  Decision Forests for Classification, Regression, Density Estimation, Manifold Learning and Semi-Supervised Learning , 2011 .

[9]  Oudom Somphone,et al.  Real-Time 3D Image Segmentation by User-Constrained Template Deformation , 2012, MICCAI.

[10]  J. Alison Noble,et al.  Registration of 3D Fetal Brain US and MRI , 2012, MICCAI.

[11]  Dorin Comaniciu,et al.  AutoMPR: Automatic detection of standard planes in 3D echocardiography , 2008, 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[12]  Nicholas Ayache,et al.  Rigid registration of 3-D ultrasound with MR images: a new approach combining intensity and gradient information , 2001, IEEE Transactions on Medical Imaging.

[13]  I. Timor-Tritsch,et al.  Sonographic examination of the fetal central nervous system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’ , 2007, Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology.