Login now to access Regular content available to all registered users.
Abstract
For Discussant: Click here to Download of Summary Slides (PPT)
Hybrid Dual Energy (HDE) Fluoroscopy for Real-Time Tracking of Lung Tumors
Purpose/Objectives: Dual energy (DE) imaging uses a combination of high and low energy kV planar x-ray images to remove bone and produce soft tissue enhanced images. Previous studies have shown that DE imaging improves the localization of lung tumors on planar kV x-ray images, compared to single energy imaging. Thus, combining DE imaging with fluoroscopy may enable real-time tumor motion tracking. However, such an approach would require a “fast switching” x-ray generator (FSDE) to produce alternating high and low energy x-ray images. We describe here a Hybrid DE (HDE) approach that can be used with existing imaging hardware to enhance lung tumor visualization for real-time motion tracking.
Materials/Methods: The HDE algorithm uses a single image set obtained prior to intra-fraction imaging to provide bone suppression on subsequent fluoroscopic images. Briefly, a 60 kVp and 120 kVp image set are obtained using respiratory gating, and are combined to produce a bone-weighted image. This image highlights the ribs and other skeletal bones, and suppresses the surrounding soft tissue. Next a standard fluoroscopic image sequence is obtained. Logarithmic subtraction is performed on a frame-by-frame basis using the previously obtained bone-weighted image. This subtraction results in a fluoroscopic image sequence in which the ribs are suppressed, providing improved visualization of the tumor. To validate this approach, both high and low energy fluoroscopic image sets were obtained on 3 patients (total of 9 fluoroscopic sets) using an on-board imager. To simulate FSDE, the high and low energy fluoroscopic images were aligned based on respiratory phase, and frame-by-frame subtraction was performed to produce a soft tissue fluoroscopic image set. HDE fluoroscopy sequences were produced using only the high-energy sequence and a single bone-weighted image. A template-based motion-tracking algorithm was used on both FSDE and HDE image sequences and the tracking coordinates, as well as the peak-to-side lobe ratio (PSR - a quantitative measure of the template match), were compared.
Results: A total of 1304 fluoroscopic frames were evaluated and there was no significant difference in the location of the tumor center using HDE vs. FSDE fluoroscopy. The average differences in the tumor centroid coordinates were -0.05 +/- 0.29 mm and 0.02 +/- 0.27 mm in the x- and y-directions, respectively. The absolute maximum differences in x-y coordinates were 0.74 mm and 1.17 mm, respectively. The PSR values were also comparable with average values of 3.63 +/- 0.80 vs. 3.62 +/- 0.84 (p=0.36) for FSDE and HDE fluoroscopy, respectively.
Conclusions: The HDE approach provides improved visualization of lung tumors on fluoroscopic imaging without the need for any additional hardware. Moreover, HDE provides a means for real-time motion tracking of lung tumors, producing results that are comparable to FSDE fluoroscopy.
Hybrid Dual Energy (HDE) Fluoroscopy for Real-Time Tracking of Lung Tumors
Purpose/Objectives: Dual energy (DE) imaging uses a combination of high and low energy kV planar x-ray images to remove bone and produce soft tissue enhanced images. Previous studies have shown that DE imaging improves the localization of lung tumors on planar kV x-ray images, compared to single energy imaging. Thus, combining DE imaging with fluoroscopy may enable real-time tumor motion tracking. However, such an approach would require a “fast switching” x-ray generator (FSDE) to produce alternating high and low energy x-ray images. We describe here a Hybrid DE (HDE) approach that can be used with existing imaging hardware to enhance lung tumor visualization for real-time motion tracking.
Materials/Methods: The HDE algorithm uses a single image set obtained prior to intra-fraction imaging to provide bone suppression on subsequent fluoroscopic images. Briefly, a 60 kVp and 120 kVp image set are obtained using respiratory gating, and are combined to produce a bone-weighted image. This image highlights the ribs and other skeletal bones, and suppresses the surrounding soft tissue. Next a standard fluoroscopic image sequence is obtained. Logarithmic subtraction is performed on a frame-by-frame basis using the previously obtained bone-weighted image. This subtraction results in a fluoroscopic image sequence in which the ribs are suppressed, providing improved visualization of the tumor. To validate this approach, both high and low energy fluoroscopic image sets were obtained on 3 patients (total of 9 fluoroscopic sets) using an on-board imager. To simulate FSDE, the high and low energy fluoroscopic images were aligned based on respiratory phase, and frame-by-frame subtraction was performed to produce a soft tissue fluoroscopic image set. HDE fluoroscopy sequences were produced using only the high-energy sequence and a single bone-weighted image. A template-based motion-tracking algorithm was used on both FSDE and HDE image sequences and the tracking coordinates, as well as the peak-to-side lobe ratio (PSR - a quantitative measure of the template match), were compared.
Results: A total of 1304 fluoroscopic frames were evaluated and there was no significant difference in the location of the tumor center using HDE vs. FSDE fluoroscopy. The average differences in the tumor centroid coordinates were -0.05 +/- 0.29 mm and 0.02 +/- 0.27 mm in the x- and y-directions, respectively. The absolute maximum differences in x-y coordinates were 0.74 mm and 1.17 mm, respectively. The PSR values were also comparable with average values of 3.63 +/- 0.80 vs. 3.62 +/- 0.84 (p=0.36) for FSDE and HDE fluoroscopy, respectively.
Conclusions: The HDE approach provides improved visualization of lung tumors on fluoroscopic imaging without the need for any additional hardware. Moreover, HDE provides a means for real-time motion tracking of lung tumors, producing results that are comparable to FSDE fluoroscopy.
For Discussant: Click here to Download of Summary Slides (PPT)
Hybrid Dual Energy (HDE) Fluoroscopy for Real-Time Tracking of Lung Tumors
Purpose/Objectives: Dual energy (DE) imaging uses a combination of high and low energy kV planar x-ray images to remove bone and produce soft tissue enhanced images. Previous studies have shown that DE imaging improves the localization of lung tumors on planar kV x-ray images, compared to single energy imaging. Thus, combining DE imaging with fluoroscopy may enable real-time tumor motion tracking. However, such an approach would require a “fast switching” x-ray generator (FSDE) to produce alternating high and low energy x-ray images. We describe here a Hybrid DE (HDE) approach that can be used with existing imaging hardware to enhance lung tumor visualization for real-time motion tracking.
Materials/Methods: The HDE algorithm uses a single image set obtained prior to intra-fraction imaging to provide bone suppression on subsequent fluoroscopic images. Briefly, a 60 kVp and 120 kVp image set are obtained using respiratory gating, and are combined to produce a bone-weighted image. This image highlights the ribs and other skeletal bones, and suppresses the surrounding soft tissue. Next a standard fluoroscopic image sequence is obtained. Logarithmic subtraction is performed on a frame-by-frame basis using the previously obtained bone-weighted image. This subtraction results in a fluoroscopic image sequence in which the ribs are suppressed, providing improved visualization of the tumor. To validate this approach, both high and low energy fluoroscopic image sets were obtained on 3 patients (total of 9 fluoroscopic sets) using an on-board imager. To simulate FSDE, the high and low energy fluoroscopic images were aligned based on respiratory phase, and frame-by-frame subtraction was performed to produce a soft tissue fluoroscopic image set. HDE fluoroscopy sequences were produced using only the high-energy sequence and a single bone-weighted image. A template-based motion-tracking algorithm was used on both FSDE and HDE image sequences and the tracking coordinates, as well as the peak-to-side lobe ratio (PSR - a quantitative measure of the template match), were compared.
Results: A total of 1304 fluoroscopic frames were evaluated and there was no significant difference in the location of the tumor center using HDE vs. FSDE fluoroscopy. The average differences in the tumor centroid coordinates were -0.05 +/- 0.29 mm and 0.02 +/- 0.27 mm in the x- and y-directions, respectively. The absolute maximum differences in x-y coordinates were 0.74 mm and 1.17 mm, respectively. The PSR values were also comparable with average values of 3.63 +/- 0.80 vs. 3.62 +/- 0.84 (p=0.36) for FSDE and HDE fluoroscopy, respectively.
Conclusions: The HDE approach provides improved visualization of lung tumors on fluoroscopic imaging without the need for any additional hardware. Moreover, HDE provides a means for real-time motion tracking of lung tumors, producing results that are comparable to FSDE fluoroscopy.
Hybrid Dual Energy (HDE) Fluoroscopy for Real-Time Tracking of Lung Tumors
Purpose/Objectives: Dual energy (DE) imaging uses a combination of high and low energy kV planar x-ray images to remove bone and produce soft tissue enhanced images. Previous studies have shown that DE imaging improves the localization of lung tumors on planar kV x-ray images, compared to single energy imaging. Thus, combining DE imaging with fluoroscopy may enable real-time tumor motion tracking. However, such an approach would require a “fast switching” x-ray generator (FSDE) to produce alternating high and low energy x-ray images. We describe here a Hybrid DE (HDE) approach that can be used with existing imaging hardware to enhance lung tumor visualization for real-time motion tracking.
Materials/Methods: The HDE algorithm uses a single image set obtained prior to intra-fraction imaging to provide bone suppression on subsequent fluoroscopic images. Briefly, a 60 kVp and 120 kVp image set are obtained using respiratory gating, and are combined to produce a bone-weighted image. This image highlights the ribs and other skeletal bones, and suppresses the surrounding soft tissue. Next a standard fluoroscopic image sequence is obtained. Logarithmic subtraction is performed on a frame-by-frame basis using the previously obtained bone-weighted image. This subtraction results in a fluoroscopic image sequence in which the ribs are suppressed, providing improved visualization of the tumor. To validate this approach, both high and low energy fluoroscopic image sets were obtained on 3 patients (total of 9 fluoroscopic sets) using an on-board imager. To simulate FSDE, the high and low energy fluoroscopic images were aligned based on respiratory phase, and frame-by-frame subtraction was performed to produce a soft tissue fluoroscopic image set. HDE fluoroscopy sequences were produced using only the high-energy sequence and a single bone-weighted image. A template-based motion-tracking algorithm was used on both FSDE and HDE image sequences and the tracking coordinates, as well as the peak-to-side lobe ratio (PSR - a quantitative measure of the template match), were compared.
Results: A total of 1304 fluoroscopic frames were evaluated and there was no significant difference in the location of the tumor center using HDE vs. FSDE fluoroscopy. The average differences in the tumor centroid coordinates were -0.05 +/- 0.29 mm and 0.02 +/- 0.27 mm in the x- and y-directions, respectively. The absolute maximum differences in x-y coordinates were 0.74 mm and 1.17 mm, respectively. The PSR values were also comparable with average values of 3.63 +/- 0.80 vs. 3.62 +/- 0.84 (p=0.36) for FSDE and HDE fluoroscopy, respectively.
Conclusions: The HDE approach provides improved visualization of lung tumors on fluoroscopic imaging without the need for any additional hardware. Moreover, HDE provides a means for real-time motion tracking of lung tumors, producing results that are comparable to FSDE fluoroscopy.
{{ help_message }}
{{filter}}