Deep Learning Methods for Sparse-View 3D X-ray Computed Tomography
| When |
Apr 01, 2025
from 01:00 to 02:00 |
|---|---|
| Where | M7 101 |
| Attendees |
Romain Vo |
| Add event to calendar |
vCal
iCal
|
Romain VO
Title: Deep Learning Methods for Sparse-View 3D X-ray Computed Tomography
X-ray computed tomography (CT) involves the reconstruction of the 3D image of an object from a set of measurements called radiographs. It is an essential imaging technique in the medical field, as well as the non-destructive testing of industrial components. In these two applications, there is a common need to produce reliable and high-quality images using a minimal number of radiographs. Reducing the number of measurements is done to limit the patient's dose or reduce the acquisition time, making it compatible with online production control. Unfortunately, reducing the number of projections leads to the appearance of artifacts in the reconstructed image, significantly affecting its quality. During my thesis, I focused on building deep learning techniques to tackle the problem of sparse-view 3D X-ray CT. The seminar today will cover three main areas:
1. An overview of the current state-of-the-art deep learning methods for sparse-view CT and linear imaging inverse problems.
2. The development of a memory-efficient procedure to reconstruct high-quality 3D images from minimal projections,
3. An evaluation framework that goes beyond standard distortion metrics to assess the quality of reconstructed images, ensuring they are suitable for specific applications using the observer framework.
Website: https://romainvo.github.io/
In Room M7 101, 1st floor, Monod campus, ENSL.