I am a PhD candidate in the Computational Imaging Group at the Centrum Wiskunde & Informatica (CWI) in Amsterdam advised by Prof. Joost Batenburg.
Research: My research focuses on developing deep learning approaches to improve image quality in practical scenarios where very little training data is available.
November 2021: Our article Tomosipo: Fast, flexible, and convenient 3D tomography for complex scanning geometries in Python has been published in Optics Express.
November 2021: Visiting the group of Alexandra Pacureanu at the ID16A beamline at the ESRF. Noise2Inverse shows exciting results on reconstructed tomographic images of real neural networks!
July 2021: I won third place in the Fujitsu Multi-node GPU Challenge. My submission using PyTorch and Horovod is published on GitHub.
July 2021: I was invited to give a talk in the webinar on AI applied to X-ray and synchrotron techniques organized by the ESRF. The recording can be found on YouTube.
June 2021: Our article Deep Denoising for Multi-Dimensional Synchrotron X-Ray Tomography Without High-Quality Reference Data has been published in Scientific Reports.
May 2021: A new version of msd_pytorch has been released including support for 3D convolutions. The CUDA implementation was ported from 2D by Ryan Pollit as a master student project (huge thanks!).
May 2021: I was selected to give a talk at the final of the PhD prize competition organized by the Dutch royal mathematical society (KWG).
August 2020: I won the best poster prize at the Mathematics of Machine Learning symposium of the London Mathematical Society and the University of Bath. Poster title: Noise2Inverse: Deep tomographic denoising without high-quality target data. (More)