Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Occiput.io - Tomographic Vision

Software for volumetric imaging: PET, MRI, SPECT, X-RAY CT. Reconstruct from the raw imaging data, register, segment, visualize.
by

Stefano Pedemonte

on 1 December 2015

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Occiput.io - Tomographic Vision

- UNIFIED
- WEB-ORIENTED
SPECT
PET
X-RAY CT
MR
PET-MR
share
collaborate
distribute

- HIGH PERFORMANCE
SOFTWARE FOR VOLUMETRIC IMAGING
Imaging primitives at the bottom
Design on principle
vNAV
275 ms
Reg.
80 ms
MPRAGE
readout
inv. pulse
TR gap
vNAV
275 ms
Reg.
80 ms
MPRAGE
readout
inv. pulse
TI gap
TI gap
TR gap
vNAV
275 ms
Reg.
80 ms
MPRAGE
readout
inv. pulse
TI gap
..
MOTION EVENT DISCRIMINATION
ACTIVITY
MOTION
COUNTS
Registration
Registration
REGISTRATION
MOTION
UNCERTAINTY
RIEMANNIAN MANIFOLD TOMOGRAPHY
Contributors:
Stefano Pedemonte
Niccolò Fuin
David Izquierdo
Ciprian Catana
Spencer Bowen

This research was supported by:

NIH NCRR
NIH NIBIB
TEKES
The Lundbeck Foundation
NVIDIA Corporation

Images:
Scanner picture courtesy of Siemens and Philips
REGISTRATION
GPU primitives for volumetric imaging
SPECT
PET
X-Ray
CT
- Ray-tracing:




- Transformation:


- Stancil Operations


- FFT - IFFT

- transmission
- cone beam
- parallel beam
- SPECT
- PET
- affine
- B-splines
- uniform
- spatially variant
Organization of the code


How to install


How to integrate













GPU High Performance Computing
PIPELINE
MODEL-BASED
1) install NiftyRec - CMake

2)
python setup.py build install
or
pip install occiput
- Python

- Matlab ray-tracers

- C/C++ ray-tracers *
*Kitware RTK uses NiftyRec for CT reconstruction
NiftyRec libraries

UCL
2010-2013
Harvard University
2013-2015
.
more than 10K downloads
GATING
- load motion information
- create computational graph for motion correction
- load gating information
- create computational graph for gated reconstruction
DisplayNode
Web-based interactive visualization
IMAGE QUALITY
Comparison between Occiput and Siemens' E7-tools
Normalization:
Yes
Attenuation: No
Scatter correction: No
Randoms correction: No
Number of MLEM iterations: 40
CYLINDER PHANTOM
CYLINDER PHANTOM
Normalization:
Yes
Attenuation: Yes
Scatter correction: Yes
Randoms correction: Yes
Number of MLEM iterations: 40
DeRENZO PHANTOM
OCCIPUT
E7-TOOLS
NATIVE GEOMETRY
OSEM RECONSTRUCTION CODE
RECONSTRUCTION USING iLang
BRAIN STUDY
OCCIPUT
E7-TOOLS
Saggital plane
Transverse plane
Sagittal axis
Coronal axis
Blue - OCCIPUT
Red - E7-TOOLS
Normalization:
Yes
Attenuation:
Yes
Scatter correction: Yes
Randoms correction: Yes
Number of MLEM iterations: 40
Normalization:
Yes
Attenuation: No
Scatter correction: No
Randoms correction: No
Number of MLEM iterations: 40
BLUE: OCCIPUT
RED: E7-TOOLS
BLUE: OCCIPUT
RED: E7-TOOLS
SPECT

MR

PET
Description of the imaging system
pure Python
pure Python
- Python
- plugins for listmode streaming
and index to LOR
1 - MAXIMUM PROBABILITY

-
ALTERNATING DIRECTION METHOD OF MULTIPLIERS
- PRECONDITIONED GRADIENT DESCENT
- LBFGS (QUASI NEWTON)
- NEWTON
2 - POSTERIOR SAMPLING


HAMILTONIAN MARKOV CHAIN MONTE CARLO

INFERENCE
Full transcript