Site-specific volumetric analysis of lung tumour motion

Eric W. Pepin, Huanmei Wu, George A. Sandison, Mark Langer, Hiroki Shirato

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


The treatment of lung cancer with radiation therapy is hindered by respiratory motion. Real-time adjustments to compensate for this motion are hampered by mechanical system latencies and imaging-rate restrictions. To better understand tumour motion behaviour for adaptive image-guided radiation therapy of lung cancer, the volume of a tumour's motion space was investigated. Motion data were collected by tracking an implanted fiducial using fluoroscopy at 30 Hz during treatment sessions. A total of 637 treatment fractions from 31 tumours were used in this study. For each fraction, data points collected from three consecutive breathing cycles were used to identify instantaneous tumour location. A convex hull was created over these data points, defining the tumour motion envelope. The study sought a correlation between the tumour location in the lung and the convex hull's volume and shape. It was found that tumours located in the upper apex had smaller motion envelopes (<50 mm3), whereas tumours located near the chest wall or diaphragm had larger envelopes (>70 mm3). Tumours attached to fixed anatomical structures had small motion spaces. Three general shapes described the tumour motion envelopes: 50% of motion envelopes enclosed largely 1D oscillation, 38% enclosed an ellipsoid path, 6% enclosed an arced path and 6% were of hybrid shape. This location-space correlation suggests it may be useful in developing a predictive model, but more work needs to be done to verify it.

Original languageEnglish (US)
Pages (from-to)3325-3337
Number of pages13
JournalPhysics in medicine and biology
Issue number12
StatePublished - 2010
Externally publishedYes

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Site-specific volumetric analysis of lung tumour motion'. Together they form a unique fingerprint.

Cite this