This Article Abstract Full Text Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ahn, S Articles by Choi, E Search for Related Content PubMed PubMed Citation Articles by Ahn, S Articles by Choi, E

A feasibility study on the prediction of tumour location in the lung from skin motion

¹ Department of Radiation Oncology, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap-dong Songpa-gu Seoul and ² Department of Physics, Ewha Womans University, Seoul, Korea

View larger version (14K): [in a new window]   Figure 1. Scheme for deducing the correlation between skin markers and tumour movements: observing skin markers and internal tumour synchronously, and then inferring the correlation between them. AP, anterior–posterior.

View larger version (52K): [in a new window]   Figure 2. Fluoroscopic images of skin markers and organs (a) in the anterior–posterior view and (b) in the lateral view: the locations of skin and organ were determined from several points on the individual skin markers and organs during normal breathing.

View larger version (72K): [in a new window]   Figure 3. Images of skin markers and internal organs during (a) inspiration and (b) expiration, and (c) their difference, in the lateral view.

View larger version (42K): [in a new window]   Figure 4. Respiratory motion patterns of skin and tumour: motion patterns of skin markers and tumour (a) in the x-axis (right upper lung fields in the anterior–posterior (AP) view), (b) in the y-axis (right upper lung fields in the AP view) and (c) in the z-axis (lower lung fields in the lateral view). (d) Motion patterns in the y-axis (right upper lung fields in the AP view): when skin makers began to move, a short delay in motion was caused by respiratory tumour motion. (e) Motion patterns in the y-axis (left upper lung fields in the AP view): when distortions were caused by sudden changes in period, amplitude, or phase of respiration, skin and tumour experienced the same distortions, and thus, remained correlated. x-axis, right–left direction of the body; y-axis, superior–inferior direction; z-axis, anterior–posterior direction.

View larger version (19K): [in a new window]   Figure 5. Correlation between skin and tumour movements: correlation coefficient of (a) the best case was 0.94 (upper lung fields in the lateral view), while that of (b) the worst case was 0.41 (left upper lung fields in the anterior–posterior view).

View larger version (46K): [in a new window]   Figure 6. Prediction of internal tumour location from external skin markers, and the uncertainties of tumour position due to movement during breathing before prediction (actual tumour motion) and after prediction (corrected tumour motion, which were the difference between actual tumour motion and predicted motion): motions (a) in the y-axis (right lower lung fields in the anterior–posterior (AP) view) and (b) in the y-axis (lower lung fields in the lateral view) were predicted by an amplitude fitting, and motions (c) in the y-axis (right lower lung fields in the AP view) by a phase-corrected fitting. (d) Prediction with remote skin markers: from skin motion in the z-axis in the lateral view, tumour motion in the y-axis in the AP view was predicted (lower lung fields in the lateral view). The uncertainties of tumour position were reduced (a) from 2.26 cm to 0.92 cm, (b) from 2.26 cm to 1.05 cm, (c) from 1.78 cm to 0.77 cm, and (d) 1.74 cm to 0.72 cm. x-axis, right–left direction of the body; y-axis, superior–inferior direction; z-axis, anterior–posterior direction.