600'446 projects

1
600.446 projects

• Radiation Oncology

2
Close Loop Adaptive Radiation Therapy
RTP
Analysis of early treatment information
3
Image Calendar to Manage Adaptive Workflow
4
Example of tumor shrinkage during the course of
radiation treatment
5
Infrastructure at JHMI TODAY
Exists Future
Paper Charts
Radiology, Nuc Med
DICOM
Radiation Oncology
Portal Image
jpg pdf
Treatment Images
DICOM
RT Applications
Linear Accelerator
6
Infrastructure at JHMI TOMORROW
Need Physician friendly ART Tools
Radiology, Nuc Med
DICOM
Radiation Oncology
Treatment Images
DICOM
RT Applications
Linear Accelerator
7
PixelMed Java DICOM Tools
8
Web Tool Development

• Current
• Treatment plan review initially
• JPG and PDF exports from TPS
• Futures
• Electronic signature
• DICOM RT Viewer
• Adaptive plan review
• Risk
• Commercial solution availability

9
Tools for Adaptive Radiation Therapy(Stick with
anatomical for project)

• Project 1 Infra-structure Development
• Implement a web-based distribution environment
  that enables the clinical personnel to receive,
  review, and act on information for adaptive
  radiation therapy begin with a simple model of
  repeat radiographic and tomographic image review
  and analysis
• Project 2 Image Analysis Tools
• 2D/2D 2D/3D 3D/3D 3D/4D 4D/4D registrations
• Deformable organ registration
• Analysis presentation (identify pertinent
  information in few minutes)
• Cine loop motion display
• Data plots of organ motion

10
Project 1 ART Infrastructure Development

• Integrate tools and infra-structure to facilitate
  adaptive radiation therapy
• Radiation Oncology Programmatic Initiatives
  (internal and industrial funding)
• What students will do
• Identify efficient presentation of treatment plan
  information over web
• Develop Java applet for DICOM (RT) object display
  on web
• Portal images to start (scope to remain
  manageable)
• Support for PACS like image data access
• Identify user (physician) requirements to find
  optimal GUI for workflow and information
  presentation
• Deliverables
• Participate to build a large software suite to
  support adaptive radiation therapy workflow
• Size group preferably 2 or 3 per project
• Skills Needed C/C, Java, Web based development
• Mentoring
• Todd McNutt, Ray Gaudette, John Wong, Eric Ford,
  Erik Tryggestad, Ted DeWeese

11
Project 2 ART Image Analysis

• Integrate tools and infra-structure to facilitate
  adaptive radiation therapy
• Radiation Oncology Programmatic Initiatives
  (internal and industrial funding)
• What students will do
• Integrate image registration tools for analysis
  of 2D portal images and 3D CT scans
• Develop tools to present motion analysis in an
  efficient manner
• Cine loops
• Motion plots
• Deliverables
• Suite of image registration and analysis tools
  to support adaptive radiation therapy workflow
• Size group preferably 2 or 3 per project
• Skills Needed Java, MATLAB, C/C,
• Mentoring
• John Wong, Todd McNutt, Eric Ford, Erik
  Tryggestad, Ted DeWeese

12
4D Analysis Organ Models

• 4D CT of the whole breathing cycle taken prior to
  treatment
• Organ propagation of lung and automated
  deformable volumetric registration

Images courtesy MD Anderson Cancer Center
Images courtesy MD Anderson Cancer Center
13
Secondary dataset with primary IMRT
beam arrangement
Primary dataset
Model-based segmentation
Deformable registration
Dose warping
Secondary dose deformed back to primary plan
Images Courtesy William Beaumont Hospital
14
Project 3 4D Analysis of data

• Analyze motion characteristic of several patients
  and assess treatment quality
• Radiation Oncology Programmatic Initiatives
  (internal and industrial funding)
• What students will do
• Evaluate several patients course of treatment
  with the available images
• Identify dosimetric impact of the motion
• Identify characteristics of target changes (tumor
  response)
• Seek efficient mechanism for analysis
• Deliverables
• Results of analysis
• Identification of a practical efficient analysis
  process
• Size group preferably 2 or 3 per project
• Skills Needed Java, C/C, Pinnacle
• Mentoring
• John Wong, Todd McNutt, Eric Ford, Erik
  Tryggestad, Ted DeWeese

15
Focused x-ray dosimetry
4 cm
12/28/05
16
Focused x-ray Dosimetry

• Implement a convolution/pencil beam summation
  algorithm for dose calculation
• NIH R01 project (PI is John Wong, Rad Onc),
  potential for summer support MS project
• What students will do
• Develop the pencil beam summation algorithm that
  works on measured data
• Integrate the calculated dose distribution for
  display on a commercial treatment planning system
• Perform phantom validation studies
• Publication
• Deliverables
• Working dose calculation algorithms
  participation in experiments reports
• Size group preferably 2 or 3 (1 may still be
  helpful)
• Skills Needed MATLAB, C/C, math, experimental
  skills
• Mentoring
• John Wong, Todd McNutt, Erik Tryggestad, Howard
  Deng

17
Cone Beam CT
Transverse orientation
Coronal orientation
18
Cone beam CT Reconstruction

• Implement a cone beam CT reconstruction algorithm
  for the small animal irradiation system.
• NIH R01 project (PI is John Wong, Rad Onc),
  potential for summer support MS project
• What students will do
• Develop image processing and reconstruction
  software tools
• Develop x-ray image acquisition control based on
  a CCD camera system and later on a flat panel
  detector (if time permits)
• Perform phantom and animal imaging studies
• Publication
• Deliverables
• Working reconstruction tools participation in
  experiments reports
• Size group preferably 2 or 3 (1 may still be
  helpful)
• Skills Needed MATLAB, C/C, math, experimental
  skills
• Mentoring
• John Wong, Eric Ford, Peter Kazanzedis, (Gerry
  Prince?)

19
Fiducials for registration
20
The oxygenation / radiation relation a pilot
PET study

• Background Low-oxygen regions are resistant to
  radiation. Spatial distribution can be imaged
  with positron emission tomography (PET)
• Goal Understand interaction of radiation and
  oxygenation to improve radiation treatment.
• Pilot study for future NIH funding
• Tasks
• Develop image registration algorithm to correlate
  high-resolution histology data to
  lower-resolution PET measurements.
• Define methods for delineating hypoxic regions
  and quantifying dynamic changes.
• Work with a kinematic model of tracer uptake and
  compare with static imaging.
• Deliverables
• Registration software PET-based quantification
  methods
• Size group 3-4 people
• Skills Needed Image analysis programming
• Mentoring
• Eric Ford, PhD, Department of Radiation Oncology
• eric.ford_at_jhmi.edu 410-502-1477