In March 2019, Wietske started her PhD project about developing a spatiotemporal brain atlas of the developing embryonic brain based on ultrasound imaging using deep learning. Her project is a collaboration between the Periconception Epidemiology group of the Department of Obstetrics and Gynaecology and the Biomedical Imaging Group Rotterdam (BIGR) of the Department of Radiology and Nuclear Medicine.
After her PhD defence (expected in March 2024), she will continue as a postdoctoral researcher. During her PhD project, Wietske developed different deep learning-based methods to automatically monitor the growth and development of the embryonic and fetal brain. All methods make use of 3D first-trimester ultrasound. She proposed a framework to simultaneously segment and spatially the embryo, using a set of references that were manually segmented and aligned. These two tasks form the basis for automatic monitoring of growth and development: placing the embryo in a standard orientation enables derivation of the standard planes, and the segmentation of the embryo provides the embryonic volume. The advantage of the proposed framework is that after alignment, any standard plane and associated biometric measurements can easily be identified. Besides alignment, Wietske focussed on measuring the embryonic volume and embryonic head volume automatically using deep learning.
Finally, she developed a deep learning approach to create a 4D spatiotemporal atlas of the embryonic brain. This atlas consists of a template per gestational day and is defined as the mean morphology of the brain. The resulting atlas gives a detailed picture of normal embryonic and fetal brain development and was used to quantitatively assess the influence of a high maternal BMI during early pregnancy. During her postdoctoral training, Wietske will focus on modelling embryonic, fetal and placenta development to detect adverse growth and congenital anomalies and get insight into the influence of lifestyle behaviour and maternal and paternal conditions.
Looking for an interesting internship or thesis project? Contact me!
Education and career
- Mar 2019 – Present: PhD candidate at Erasmus MC Project: Spatio-temporal modelling of embryonic brain development. Collaboration between the Biomedical Imaging Group Rotterdam (BIGR, part of the Department of Radiology and Nuclear Medicine), and Periconceptional Epidemiology (part of Department of Obstetrics and Gynaecology).
- 2016 - 2018: University of Twente Master Applied Mathematics, grade 8/10 on average. Chair Applied Analysis Topics: inverse problems, imaging, (functional) analysis, optimization, dynamical systems, partial differential equations, bifurcation theory.
- 2012 - 2016: University of Twente Bachelor Applied Mathematics, grade 7.8/10 on average. Minor Philosophy of Science and Technology. Contributed to publication in Operations Research for Health Care.
Computational methods for the analysis of early-pregnancy brain ultrasonography: a systematic review. W.A.P. Bastiaansen et al. eBioMedicine, 2023 89: 1 - 22.
Multi-Atlas Segmentation and Spatial Alignment of the Human Embryo in First Trimester 3D Ultrasound. W.A.P. Bastiaansen et al. Machine Learning for Biomedical Imaging, PIPPI 2021 special issue, 2022 1(1): 1-31.
- Prenatal Image Analysis.
- Medical Image Analysis.
- First Trimester Ultrasound.
- Image Registration.
- Embryonic and Fetal Brain Development.
- Machine learning.
- Deep learning.
- Artificial Intelligence.