2006 IGT Workshop Radiation Therapy Breakout

Back to workshop agenda

Questions to be answered in the report-out session

• Identify 3 main challenges in this area
• Identify 3 specific problems that can be solved by a collaborative effort between academic and industry partners
• Identify a problem that the NCIGT can help address in the next year

Minutes

User:Shoge RAD breakout notes:

• Review of BWH program
  • seed placement
  • treated 472 men, 2 cases a week
  • mri guided biopsy program
  • mr is used throughout (both) program(s)
  • will shift to 3T in a few weeks

• Typical proceedure
  • segment: peripheral zone (CTV), urethra, rectal (180° angular coverage)
  • two 18g needles: preloaded w/ sources/spacers,
    • location conf, then drop the seeds
  • Joy: guidance allows plan modification during treatment
    • not uncommon to add additional needles, seed points
  • review deviations, to confirm proper coverage:
    • goal: 100% peripheral zone coverage; if miss, miss in a particular region.
    • harder to visualize seed placement as treatment proceeds (signal voids from needle track)

• 3D data set
  • body coil
  • extern phase array (cardiac coil?)
  • T1/T2, Dyn Contrast Enhanced, DWI (line scan), MRSI,
  • Challenge: how to combine all of this data.

• corregistration: seeds in MR compare w/ CT,
  • reg algorithm uses fiducials (the seeds).
  • distortion? not much
  • Would like to ellliminate the CT from the proceedure.
  • Problem: 100 seeds from CT, can we accurately register with 50 seeds visible in MR.

• (Brac) Outcome validation
  • O.R. time, staff costs
  • patient experience
  • 'controls' is a challege. rectal toxicity data; compared to similarly matched pool of patients (temporally correlated); manage side-effects;
  • patient reported scores on Quality of Life are higher than ultrasound guided placement.
    • procedure time is a bit longer
    • shortest is between 3-4 hours (multi disciplinary team)
    • MR set up time, anathesia takes 1 to 1 1/2 hour of that time.

• Biopsy
  • register 1.5 T image from weeks ago, 0.5 T image from 5 min before proceedure, w/ real-time image during
  • target range is within 2-3 mm
  • segmentation is keystone; manual segmentation; regisitration is straightforward.

• review of case history(s)
  • prostate cancer is slow growing
  • grading? staging?

• Imaging
  • structure
  • guidance / adjustment
• imaging challenges
  • volume based
  • Partial information based
  • adjustments (human vs machine)

• challenges:
  • have MR scanning driven by location of needle under discussion by all MDs during procedure
  • training seed placement based on biopsy tests
    • seed registration in mri
  • multimodal integration and validation of targets
  • software interfaces
    • API hooks into internal commercial systems
    • current image analysis SW is inadequate for multiple images
    • impossible to get manual segmentation of all images; semi-automated segmentation... calibration of algorithms (absolute accuracy)
  • Closed bore (3T) environment
    • integration of robots into MRT

---

• national repository of data %BR% need multi-modality images of same patient (constant geometry)
• during implants: update/adjust dosimetry for optimal delivery

Summary Report

Breakout report presentation (Noby Hata)