Literature DB >> 26215586

Automatic detection of patient identification and positioning errors in radiation therapy treatment using 3-dimensional setup images.

Shyam S Jani1, Daniel A Low2, James M Lamb2.   

Abstract

PURPOSE: To develop an automated system that detects patient identification and positioning errors between 3-dimensional computed tomography (CT) and kilovoltage CT planning images. METHODS AND MATERIALS: Planning kilovoltage CT images were collected for head and neck (H&N), pelvis, and spine treatments with corresponding 3-dimensional cone beam CT and megavoltage CT setup images from TrueBeam and TomoTherapy units, respectively. Patient identification errors were simulated by registering setup and planning images from different patients. For positioning errors, setup and planning images were misaligned by 1 to 5 cm in the 6 anatomical directions for H&N and pelvis patients. Spinal misalignments were simulated by misaligning to adjacent vertebral bodies. Image pairs were assessed using commonly used image similarity metrics as well as custom-designed metrics. Linear discriminant analysis classification models were trained and tested on the imaging datasets, and misclassification error (MCE), sensitivity, and specificity parameters were estimated using 10-fold cross-validation.
RESULTS: For patient identification, our workflow produced MCE estimates of 0.66%, 1.67%, and 0% for H&N, pelvis, and spine TomoTherapy images, respectively. Sensitivity and specificity ranged from 97.5% to 100%. MCEs of 3.5%, 2.3%, and 2.1% were obtained for TrueBeam images of the above sites, respectively, with sensitivity and specificity estimates between 95.4% and 97.7%. MCEs for 1-cm H&N/pelvis misalignments were 1.3%/5.1% and 9.1%/8.6% for TomoTherapy and TrueBeam images, respectively. Two-centimeter MCE estimates were 0.4%/1.6% and 3.1/3.2%, respectively. MCEs for vertebral body misalignments were 4.8% and 3.6% for TomoTherapy and TrueBeam images, respectively.
CONCLUSIONS: Patient identification and gross misalignment errors can be robustly and automatically detected using 3-dimensional setup images of different energies across 3 commonly treated anatomical sites.
Copyright © 2015 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.

Entities:  

Mesh:

Year:  2015        PMID: 26215586     DOI: 10.1016/j.prro.2015.06.004

Source DB:  PubMed          Journal:  Pract Radiat Oncol        ISSN: 1879-8500


  2 in total

1.  A novel method for radiotherapy patient identification using surface imaging.

Authors:  David B Wiant; Quinton Verchick; Percy Gates; Caroline L Vanderstraeten; Jacqueline M Maurer; T Lane Hayes; Han Liu; Benjamin J Sintay
Journal:  J Appl Clin Med Phys       Date:  2016-03-08       Impact factor: 2.102

2.  Offline generator for digitally reconstructed radiographs of a commercial stereoscopic radiotherapy image-guidance system.

Authors:  John A Charters; Pascal Bertram; James M Lamb
Journal:  J Appl Clin Med Phys       Date:  2022-02-03       Impact factor: 2.102

  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.