Madeleine Wyburd

Madeleine Wyburd is an Associate Member of the Department of Computer Science at the University of Oxford. Her research focuses on advancing medical imaging technologies through deep learning and computer vision, particularly in the domain of fetal ultrasound analysis. She specializes in developing algorithms for 3D ultrasound reconstruction, anatomically plausible segmentation, and automated assessment of fetal brain development.

Her work bridges computer science and clinical medicine, emphasizing applications in obstetrics and prenatal care. Key contributions include techniques like RapidVol for real-time 3D ultrasound volume reconstruction and TEDS-Net, a topology-preservation network for medical image segmentation. She also explores test-time adaptation methods to improve subcortical segmentation accuracy in fetal brain imaging.

Wyburd's research integrates interdisciplinary methodologies from biomedical engineering and algorithm design. Recent projects analyze cortical plate development in second-trimester fetuses and compare 3D ultrasound with MRI volumetric measurements to enhance clinical diagnostic reliability. Her work aims to improve prenatal care through AI-driven tools that standardize fetal biometry assessment and reduce human error in clinical workflows.

She collaborates with global institutions, as evidenced by her participation in the 34th World Congress on Ultrasound in Obstetrics and Gynecology. Her studies often address practical challenges like sparse-sampling in intrapartum ultrasound and normative brain maturation tracking up to 2 years post-birth.