An intra-fraction markerless daily lung tumor localization algorithm for EPID images

Timothy Rozario, Sergey Bereg, Weihua Mao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this work we report an intra-fractional markerless algorithm that accurately detects lung tumors on mV projections within the beam's eye view, while minimizing harmful effects such as poor soft tissue resolution, global image distortion, image blurring and scattering due to intrafraction target motion and radiation scatter.. First, we generate two sets of DRRs digitally reconstructed radiographs-background DRR without tumor and tumor only DRR from the 4D CT planning data after the tumor has been initially segmented out. Next, the composite DRR is generated by fusing the tumor DRR on the background. The composite DRR along with the matching mV projection are divided into a matrix of small tiles. The tile configuration is automatically set up such that the tumor always remains within the beam's-eye-view geometry on the composite DRR. In order to locate the tumor on the mV projection, the tumor DRR is fused at different locations on the background DRR while the tiles of the composite DRR are globally shifted. For each configuration, the composite DRR is matched with the corresponding mV projection. A simple NCC normalized cross correlation is used to compute the similarity between the composite DRR and corresponding mV projection tiles. Finally, the location of the lung tumor on the mV projection is identified based on the best match found. The algorithm was successfully tested on a dynamic chest phantom at our institution. Approximately 5700 raw images over 12 gantry angles were tested and the tumor was accurately located on every mV projection. Although, the chest phantom was created to mimic the human chest anatomy with neighboring organs, tissues and bony structures, which introduced strong signals, the maximum error reported was less that 1.6 mm while the average error reported was less than 0.7 mm.

Original languageEnglish (US)
Title of host publication8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings
PublisherAssociation for Computing Machinery, Inc
ISBN (Print)9781450334525
DOIs
StatePublished - Jul 1 2015
Event8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Corfu, Greece
Duration: Jul 1 2015Jul 3 2015

Other

Other8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015
CountryGreece
CityCorfu
Period7/1/157/3/15

Fingerprint

Tumors
Tile
Composite materials
Tissue
Scattering
Radiation
Planning
Geometry

Keywords

  • Beam's eye view
  • CBCT
  • DRR
  • Epid systems
  • Healthcare informatics
  • Lung cancer
  • Lung tumor tracking
  • Motion-tracking
  • MV projections
  • Radiotherapy
  • Tile-shifting
  • Tumor tracking

ASJC Scopus subject areas

  • Software
  • Computer Science Applications

Cite this

Rozario, T., Bereg, S., & Mao, W. (2015). An intra-fraction markerless daily lung tumor localization algorithm for EPID images. In 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings [a22] Association for Computing Machinery, Inc. https://doi.org/10.1145/2769493.2769511

An intra-fraction markerless daily lung tumor localization algorithm for EPID images. / Rozario, Timothy; Bereg, Sergey; Mao, Weihua.

8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings. Association for Computing Machinery, Inc, 2015. a22.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rozario, T, Bereg, S & Mao, W 2015, An intra-fraction markerless daily lung tumor localization algorithm for EPID images. in 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings., a22, Association for Computing Machinery, Inc, 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015, Corfu, Greece, 7/1/15. https://doi.org/10.1145/2769493.2769511
Rozario T, Bereg S, Mao W. An intra-fraction markerless daily lung tumor localization algorithm for EPID images. In 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings. Association for Computing Machinery, Inc. 2015. a22 https://doi.org/10.1145/2769493.2769511
Rozario, Timothy ; Bereg, Sergey ; Mao, Weihua. / An intra-fraction markerless daily lung tumor localization algorithm for EPID images. 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings. Association for Computing Machinery, Inc, 2015.
@inproceedings{f2bbff3ed9f64809994b06161fa7bc7c,
title = "An intra-fraction markerless daily lung tumor localization algorithm for EPID images",
abstract = "In this work we report an intra-fractional markerless algorithm that accurately detects lung tumors on mV projections within the beam's eye view, while minimizing harmful effects such as poor soft tissue resolution, global image distortion, image blurring and scattering due to intrafraction target motion and radiation scatter.. First, we generate two sets of DRRs digitally reconstructed radiographs-background DRR without tumor and tumor only DRR from the 4D CT planning data after the tumor has been initially segmented out. Next, the composite DRR is generated by fusing the tumor DRR on the background. The composite DRR along with the matching mV projection are divided into a matrix of small tiles. The tile configuration is automatically set up such that the tumor always remains within the beam's-eye-view geometry on the composite DRR. In order to locate the tumor on the mV projection, the tumor DRR is fused at different locations on the background DRR while the tiles of the composite DRR are globally shifted. For each configuration, the composite DRR is matched with the corresponding mV projection. A simple NCC normalized cross correlation is used to compute the similarity between the composite DRR and corresponding mV projection tiles. Finally, the location of the lung tumor on the mV projection is identified based on the best match found. The algorithm was successfully tested on a dynamic chest phantom at our institution. Approximately 5700 raw images over 12 gantry angles were tested and the tumor was accurately located on every mV projection. Although, the chest phantom was created to mimic the human chest anatomy with neighboring organs, tissues and bony structures, which introduced strong signals, the maximum error reported was less that 1.6 mm while the average error reported was less than 0.7 mm.",
keywords = "Beam's eye view, CBCT, DRR, Epid systems, Healthcare informatics, Lung cancer, Lung tumor tracking, Motion-tracking, MV projections, Radiotherapy, Tile-shifting, Tumor tracking",
author = "Timothy Rozario and Sergey Bereg and Weihua Mao",
year = "2015",
month = "7",
day = "1",
doi = "10.1145/2769493.2769511",
language = "English (US)",
isbn = "9781450334525",
booktitle = "8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - An intra-fraction markerless daily lung tumor localization algorithm for EPID images

AU - Rozario, Timothy

AU - Bereg, Sergey

AU - Mao, Weihua

PY - 2015/7/1

Y1 - 2015/7/1

N2 - In this work we report an intra-fractional markerless algorithm that accurately detects lung tumors on mV projections within the beam's eye view, while minimizing harmful effects such as poor soft tissue resolution, global image distortion, image blurring and scattering due to intrafraction target motion and radiation scatter.. First, we generate two sets of DRRs digitally reconstructed radiographs-background DRR without tumor and tumor only DRR from the 4D CT planning data after the tumor has been initially segmented out. Next, the composite DRR is generated by fusing the tumor DRR on the background. The composite DRR along with the matching mV projection are divided into a matrix of small tiles. The tile configuration is automatically set up such that the tumor always remains within the beam's-eye-view geometry on the composite DRR. In order to locate the tumor on the mV projection, the tumor DRR is fused at different locations on the background DRR while the tiles of the composite DRR are globally shifted. For each configuration, the composite DRR is matched with the corresponding mV projection. A simple NCC normalized cross correlation is used to compute the similarity between the composite DRR and corresponding mV projection tiles. Finally, the location of the lung tumor on the mV projection is identified based on the best match found. The algorithm was successfully tested on a dynamic chest phantom at our institution. Approximately 5700 raw images over 12 gantry angles were tested and the tumor was accurately located on every mV projection. Although, the chest phantom was created to mimic the human chest anatomy with neighboring organs, tissues and bony structures, which introduced strong signals, the maximum error reported was less that 1.6 mm while the average error reported was less than 0.7 mm.

AB - In this work we report an intra-fractional markerless algorithm that accurately detects lung tumors on mV projections within the beam's eye view, while minimizing harmful effects such as poor soft tissue resolution, global image distortion, image blurring and scattering due to intrafraction target motion and radiation scatter.. First, we generate two sets of DRRs digitally reconstructed radiographs-background DRR without tumor and tumor only DRR from the 4D CT planning data after the tumor has been initially segmented out. Next, the composite DRR is generated by fusing the tumor DRR on the background. The composite DRR along with the matching mV projection are divided into a matrix of small tiles. The tile configuration is automatically set up such that the tumor always remains within the beam's-eye-view geometry on the composite DRR. In order to locate the tumor on the mV projection, the tumor DRR is fused at different locations on the background DRR while the tiles of the composite DRR are globally shifted. For each configuration, the composite DRR is matched with the corresponding mV projection. A simple NCC normalized cross correlation is used to compute the similarity between the composite DRR and corresponding mV projection tiles. Finally, the location of the lung tumor on the mV projection is identified based on the best match found. The algorithm was successfully tested on a dynamic chest phantom at our institution. Approximately 5700 raw images over 12 gantry angles were tested and the tumor was accurately located on every mV projection. Although, the chest phantom was created to mimic the human chest anatomy with neighboring organs, tissues and bony structures, which introduced strong signals, the maximum error reported was less that 1.6 mm while the average error reported was less than 0.7 mm.

KW - Beam's eye view

KW - CBCT

KW - DRR

KW - Epid systems

KW - Healthcare informatics

KW - Lung cancer

KW - Lung tumor tracking

KW - Motion-tracking

KW - MV projections

KW - Radiotherapy

KW - Tile-shifting

KW - Tumor tracking

UR - http://www.scopus.com/inward/record.url?scp=84957021624&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957021624&partnerID=8YFLogxK

U2 - 10.1145/2769493.2769511

DO - 10.1145/2769493.2769511

M3 - Conference contribution

AN - SCOPUS:84957021624

SN - 9781450334525

BT - 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, PETRA 2015 - Proceedings

PB - Association for Computing Machinery, Inc

ER -