A goal of this project is to use images from 3D and 4D ultrasound scans of fetuses to build a shape model for fetal face. This will be done by utilizing an existing face shape model constructed from optical scans of newborn babies. Project is shared between 3D Craniofacial Image Research Laboratory (joint research laboratory between KU, Rigshospitalet and DTU) and DTU Compute.Description of the problem:
3D and 4D ultrasound scans are becoming popular and available in a prenatal care. 3D and 4D ultrasound provides great detail and the reconstructed scan images resemble photographs, such that the face of the fetus can be distinguished. 3D and 4D view is used in monitoring fetal devolvement, but rarely in a systematic way. The aim of this project is to investigate the possibility of using the 3D and 4D view for building the shape model of a fetal face, and how it can be used in measuring fetal development.
The idea is to utilize the existing face shape model, which was constructed from optical scans of newborn babies. This shape model may be fitted to ultrasound images, and the parameters of those fits may be used to estimate the mean fetal face and the deviations from the mean fetal face. Based on this, a model for the fetal face can be constructed.
The data for this project is provided by 3D Craniofacial Image Research Laboratory and includes scans from 70 pregnancies (18-21 weeks). Some of the supervision might be at 3D Craniofacial Image Research Laboratory.
The developed method should be implemented in matlab or python.
In collaboration with3D Craniofacial Image Research Laboratory
As a minimum 02502 Image analysis or a similar course. 02505 Medical Image Analysis or 02506 Advanced image analysis are desirable.