Johannes K. Steinweg

This paper proposes an automated pipeline for reconstructing fetal thorax MRI from motion-corrupted stacks across 21-36 weeks GA, using 3D CNNs and landmark-guided pose estimation.

The paper creates 3D MRI atlases of normal and abnormal fetal aortic arch anatomy and evaluates deep learning for automated multi-label segmentation.

The paper develops normative brain volumetric development curves for term-born infants and compares preterm infants to these curves, identifying individual deviations linked to clinical factors and neurocognitive outcomes.

The paper investigates microstructural cortex development in neonates with congenital heart disease and finds impaired oxygen delivery correlates with abnormal microstructure.

The paper introduces a fully automated pipeline for neonatal cortical surface reconstruction from MRI, achieving high accuracy in 85% of cases.

A linearly scaling groupwise method jointly estimates mean brain shape and longitudinal change across 275 neonates (35 to 44 weeks post-menstrual age), yielding a temporally consistent morphological atlas.