MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel
  1. Home
  2. Meeting Program
  3. All Sessions
  4. All Joint Imaging-Therapy SNAP Oral
  5. Online Automatic Brain Metastases Segmentation Web Platform for Stereotactic Radiosurgery

Online Automatic Brain Metastases Segmentation Web Platform for Stereotactic Radiosurgery

H Liu1*, Y Liu2 , S Stojadinovic3 , B Hrycushko4 , Z Wardak5 , S Lau6 , W Lu7 , Y Yan8 , S Jiang9 , X Zhen10 , R Timmerman11 , L Nedzi12 , X Gu13 , (1) ,Dallas, TX, (2) Sichuan University, Chengdu, ,(3) UT Southwestern Medical Center, Dallas, TX, (4) UT Southwestern Medical Center, Dallas, TX, (5) UT Southwestern Medical Center, Dallas, tx, (6) UT Southwestern Medical Center, Dallas, tx, (7) UT Southwestern Medical Center, Dallas, Texas, (8) UT Southwestern Medical Center, Dallas, Texas, (9) UT Southwestern Medical Center, Dallas, TX, (10) Southern Medical University, Guangzhou, ,(11) UT Southwestern Medical Center, Dallas, TX, (12) UT Southwestern Medical Center, Dallas, tx, (13) UT Southwestern Medical Center, Dallas, TX

Presentations

(Sunday, 7/29/2018) 4:00 PM - 4:55 PM

Room: Karl Dean Ballroom C

Purpose: To develop an online efficient automatic BrainMets segmentation tool for facilitating contouring stereotactic radiosurgery (SRS) treatment planning and treatment follow up.

Methods: An online automatic brain metastasis (BrainMets) delineation client/server platform was designed in this work, and developed based on Django web framework. The whole platform includes two mainly components: 1) a client side JavaScript 3D medical image viewer supporting DICOM and NIFIT formats; 2) a back-end server side including image preprocessing, BrainMets segmentation, false positive contour identification and removal and communication to the client. The graph user interface (GUI) provides two ways to import images: from local folder or DICOM database. The MRI images are imported through web browser and viewed in axial, sagittal and coronal orientations. The auto-segmentation workflow is divided into six steps: 1) when the format of imported images is DICOM, it will be converted to NIFIT format; 2) Remove skull through a robust learning-based MRI brain extraction system (ROBEX); 3) Delineate brainmets using EnDeepMedic-based segmentation; 4) Remove false positives via the spherical geometry characteristics of brain metastases; 5) Send segmentation results back to web client. If the segmentation result is not promising, the post-processing can be optionally performed based on some metrics. 6) Output the segmented structure into a DICOMRT format file.

Results: It took about 4 or 5 minutes to finish the entire automatic segmentation workflow, even for 50+ BrainMets, while manually contoured processing may take hours. Our tool also allow physician overview the specified contoured tumors for comparison and follow up study.

Conclusion: The developed web platform enables online automatic BrainMets segmentation. It is a convenient tool with significant work flow improvement for BrainMets treatment planning and treatment follow up.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email

Additional Links