Hooman Esfandiari

Project Highlights

High-level overview of selected projects.

Markerless Surgical Navigation.

From intraoperative RGB-D imaging to marker-less surgical navigation!

A fully intraoperative approach for automatic registration of spinal 3D models with contineus pose updates.

SpineDepth

From intraoperative RGB-D imaging to direct anatomical 3D reconstruction

A multi-modal approach that enables acquisition of large clinical data, tailored to pedicle screw placement, using RGB-D sensors and a co-calibrated high-end optical tracking system was developed.

X23D

From intraoperative sparse fluoroscopy to anatomical 3D reconstruciton!

A fully intraoperative 3D reconstruction method based on sparse, multi-view fluoroscopy data

Brain-Computer Interface

From the surgeon's brain to the operating room monitor!

A novel brain-computer interface for intraoperatively medical image control.

Surgical Process Modeling .

From human observation to surgical process modelling!

A surgical process model for open spinal surgeries.

Intraoperative X-ray Inpainting

From occluded X-rays to clean X-rays!

A deep-learning based inpainting algorithm for rectifying occluded intraoperative X-rays.

Spinal 2D-3D Registration

From intraoperative fluoroscopy to spinal registration

A comparative analysis of intraoperative 2D-3D registration for use in spinal surgical navigation.

Pedicle Screw Pose Estimation

From intraoperative fluoroscopy to implant pose

A framework for automatic segmentation and pose estimation of pedicle screws using deep learning.