Current Research


3D Coronary Arterial (CA) Tree Reconstruction from Multiple 2D Angiographic Projections
X-ray angiography is the most commonly used imaging modality for the detection of coronary stenoses due to its high spatial and temporal resolution of lumen contour and its utility to guide coronary interventions in real time. However, the high inter- and intra-observer variability in interpreting the geometry of 3D vascular structure based on multiple 2D image projections is a limitation in the accurate determination of lesion severity. This could be addressed by the 3D reconstruction of the coronary arterial (CA) tree. The automated reconstruction of 3D CA tree from 2D projections is challenging due to the existence of several imaging artifacts, such as vessel overlap, foreshortening, and most importantly respiratory and cardiac motion. Along with these artifacts, the acquisition geometry introduces the possibility of generating false vessel segments in the reconstruction. Our approach reduces the motion artifacts in angiographic projections by developing a new method for rigid and non-rigid motion correction. A novel point-cloud based approach is developed for reconstruction of 3D vessel centerlines by iteratively minimizing the reconstruction error. The performance of the proposed 3D reconstruction is evaluated using angiographic projections from 45 patients, producing average reprojection errors of < 0.1 mm for 3D centerlines reconstruction when co-registered with the parent vessels on projection planes. A comparison of the reconstructed 3D lumen surface with optical coherence tomography (OCT) measurements has been performed, showing no statistically significant difference in the luminal cross-sections reconstructed with our method, compared to OCT.
The 3D CA tree presented left contains Left Anterior Descending (LAD), Left Circumflex (LCx), and Right Coronary Artery (RCA). All of them are reconstructed from pairs of 2D angiographic projections.

3D Heart Mesh Reconstruction from Cine MR Slices
Cardiac magnetic resonance (CMR) is increasingly used for non-invasive evaluation of the myocardium, providing accurate assessments of left ventricular function, myocardial perfusion, oedema, and scar, all of which provide important inputs in clinical decision making. From the CMR images acquired using a standard clinical protocol (short axis stack+horizontal and vertical long axis slices), an accurate high-resolution 3D representation of the myocardium is generated using patient-specific 3D surface mesh reconstruction, integrating slice alignments and segmentation refinements for a spatially consistent slice arrangement, including deep learning methodology, both of which optimise consistency between long and short axis contours.
The 3D myocardial mesh presented left is reconstructed based on long and short axis 2D CMR slices.
PS: This is actually MY Heart; feel free to download.

Students Supervised


Co-supervised with Prof. Vicente Grau
Current:
  1. Yiying Wang, Wolfson College, University of Oxford, UK: Doctoral Student (DPhil); October 2021 - .
  2. Haorui He, St Hugh's College, University of Oxford, UK: Doctoral Student (DPhil); October 2020 - .
  3. Solace Hussein, St Peter's College, University of Oxford, UK: Research Project for 4th Year Engineering; October 2021 - .

Past:
  1. Thomas Gnodde, Christ Church, University of Oxford, UK: Research Project for 4th Year Engineering; October 2020 - June 2021.
  2. Hannah Smith, Medical Sciences Doctoral Training Centre, University of Oxford, UK: CDT short project; June 2020 - September 2020.
  3. Haorui He, Worcester College, University of Oxford, UK: Research Project for 4th Year Engineering; October 2019 - June 2020.