Fully Automatic Danger Zone Determination for SBRT in NSCLC

  • Tobias Fechter Department of Radiation Oncology, University Medical Center, Freiburg
  • Jose Dolz Inserm U703, Universit Lille 2, CHRU Lille, 59120 Loos
  • Alin Chirindel German Cancer Consortium (DKTK), partner site Freiburg
  • Matthias Schlachter German Cancer Consortium (DKTK), partner site Freiburg
  • Montserrat Carles Department of Radiation Oncology, University Medical Center, Freiburg
  • Sonja Adebahr German Cancer Consortium (DKTK), partner site Freiburg
  • Michael Mix Department of Nuclear Medicine, University Medical Center, Freiburg
  • Ursula Nestle Department of Radiation Oncology, University Medical Center, Freiburg

Abstract

Lung cancer is the major cause of cancer death worldwide. The most common form of lung cancer is non-small cell lung cancer(NSCLC). Stereotactic body radiation therapy (SBRT) has emerged as a good alternative to surgery in patients with peripheralstage I NSCLC, demonstrating favorable tumor control and low toxicity. Due to spatial relationship to several critical organs atrisk, SBRT of centrally located lesions is associated with more severe toxicity and requires modification in dose application andfractionation, which is currently evaluated in clinical trials. Therefore a classification of lung tumors into central or peripheralis required. In this work we present a novel, highly versatile, mulitmodality tool for tumor classification which requires no userinteraction. Furthermore the tool can automatically segment the trachea, proximal bronchial tree, mediastinum, gross target volumeand internal target volume. The proposed work is evaluated on 19 cases with different image modalities assessing segmentationquality as well as classification accuracy. Experiments showed a good segmentation quality and a classification accuracy of 95 %.These results suggest the use of the proposed tool for clinical trials to assist clinicians in their work and to fasten up the workflowin NSCLC patients treatment.
Published
2017-10-02
How to Cite
Fechter, T., Dolz, J., Chirindel, A., Schlachter, M., Carles, M., Adebahr, S., Mix, M., & Nestle, U. (2017). Fully Automatic Danger Zone Determination for SBRT in NSCLC. Journal of Radiation Oncology Informatics, 7(1), 1–27. https://doi.org/10.5166/jroi-7-1-26
Section
Articles