NetForum uses cookies to ensure that we give you the best experience on our website. If you continue to use the site, we'll assume that you are happy to receive these cookies on the NetForum website. Read about our cookies.
NetForum Community
Learn. Share. Optimize.
Log in | Sign up now | Submit content | Contact
Go to similar content

Tumor LOC software: Intent and clinical use

White Paper
Philips CT Marketing Philips Healthcare

Sasa Mutic, MS

Associate Professor

Chief, Clinical Medical Physics Section

Department of Radiation Oncology

Washington University School of Medicine

Siteman Cancer Center

St. Louis, Missouri

 

Introduction

Through the years, radiation oncology has benefited from many technological advances: Linear accelerators, computerized treatment planning, particle accelerators, CT simulation, IMRT delivery, and IGRT delivery, to name some of the important developments. One of the common threads and demands of these technologies has been patient positioning reproducibility through the simulation, treatment planning, and treatment delivery. Conventional simulators were a very useful tool as they allowed exact patient positioning reproducibility between simulation and treatment planning. With conventional simulators, we were able to reproduce patient positioning on the treatment machine and there were no additional steps required between the simulator and the treatment machine. The major limitation of conventional simulators is the inability to provide volumetric patient information for 3D and IMRT treatment planning.

 

CT scanners and CT simulators (we'll describe the difference later) on the other hand, can provide the data necessary for 3D and IMRT-based treatment plans. However, CT technology has historically been plagued with unreliable patient setup reproducibility between the scanner and treatment machine, due to the geometry of conventional CT scanners. We were not able to scan patients in optimal treatment positions. In many institutions, early solutions were to first set up a patient in the conventional simulator, and then follow with a CT scan; the final product would be a merge between conventional simulation and a CT scan in a treatment-like position. Other institutions performed only a CT scan and then modified and adjusted patient positioning on the treatment machine to improve patient comfort and reproducibility for daily treatments. However, neither of these solutions is acceptable as they lead to disagreement between treatment plans and the actual delivered dose distributions. CT scanner manufacturers have appreciated this problem and strived to address it with CT scanners or CT simulators designed specifically with radiation therapy needs in mind.

 

The ideal situation would be to develop a CT simulator which would allow easy and reproducible patient positioning between the scanner and the treatment machine, just as the conventional simulator did so well for so many years. This ideal instrument would allow comfortable patient placement in optimal treatment positions, would provide diagnostic quality CT images, and images with large field of view with high quantitative accuracy, to facilitate accurate dose calculation accuracy. The device would be a standalone system which would facilitate efficient and accurate patient imaging. It would easily interface with other treatment planning and delivery devices in the clinic. Most importantly, the device would function, as much as possible, as the conventional simulator where patients could move from simulation to treatment without any modifications in positioning or re-marking of the patient skin. Any modifications in setup can lead to random errors and inaccuracies in patient treatments.

 

If a scanner can accomplish the above listed goals, it is considered a CT simulator. If modifications are required in patient positioning, or re-marking of patient between simulation and treatment, or if there are any other obstacles in the workflow, the device then functions more like a CT scanner than a CT simulator.

 

Philips Healthcare produced a CT scanner with four important features which allow it to be considered a CT simulator. The Philips CT Big Bore has an 85 cm gantry opening, a 60 cm true field of view (FOV), respiratory correlated imaging capabilities, and dedicated simulation software which resides on the scanner control console.

 

  • Scanner bore size: The 85 cm opening is approximately the same opening as the aperture that Varian Medical Systems accelerator, equipped with tertiary MLC, creates as it rotates around a patient. Incidentally, the 85 cm opening is also found on the tomotherapy treatment machine. The size of the CT Big Bore allows exactly the same positioning between the simulator and the treatment machine.
  • FOV: The 60 cm true field of view allows visualization of full anatomy for the majority of our patient population without any compromise in image spatial integrity or quantitative accuracy. This is especially important for those patients who weigh more than approximately 250 pounds, depending on BMI. Anatomy for these patients often can not be fully visualized on conventional 50 cm true field of view images. While some scanners have an extrapolated FOV that is larger than 50 cm, it has been shown that the extrapolated field of view has inaccurate CT numbers, and more importantly, there are spatial distortions in the extrapolated portion of the images. These spatial distortion and inaccurate CT numbers can lead to serious dosimetric errors. With 60 cm true field of view, the CT Big Bore can image the vast majority of the population without any distortions or inaccuracies in CT numbers.
  • Respiratory correlated imaging: The respiratory correlated imaging and analysis software, which is located directly on the scanner control console, enables virtual fluoroscopy, and decisions regarding patient positioning and treatment, similar to what could have been done with a conventional simulator. While respiratory correlated imaging and delivery in radiation oncology as a whole are still in their infancy, this feature and its future developments will further enable The CT Big Bore to function as a CT simulator.
  • Tumor LOC software: The last component of the CT Big Bore simulator which enables it to function as a CT simulator is the most responsible for facilitating efficient workflow and accurate patient treatments. Since this tool often involves modifications in established clinical practices, and training of several members of the radiation therapy team, its abilities often seem underutilized.

 

This manuscript describes the benefits of Tumor LOC software and how it can be implemented to optimally utilize the CT Big Bore Simulator. The conclusions drawn here are largely based on over six years of experience with this software at our institution.

 Axial image with reference marks Coronal image with reference points
Axial image with reference marks
Coronal image with reference points

 

Tumor LOC software

During CT imaging for radiotherapy treatment planning, a set of reference marks should be placed on the patient so the treatment position can be later reproduced at the treatment machine. When, and in relationship to, which anatomical landmarks these reference marks are placed can be done in two different ways. We will refer to them as shift and no-shift methods.

 

Shift method-this method is based on the process where the reference marks are placed on the patient prior to the CT scan in a somewhat arbitrary location that is close to the desired treatment isocenter. The reference point placement can be based on the diagnostic workup (CT, MRI, PET, palpation, etc.) or other physician's instructions. After the CT scan, the patient can go home and images are transferred to the treatment planning workstation.

 

Later, the physician contours target volumes and determines the treatment isocenter coordinates. Shifts (distances in three directions) between the reference marks placed on the CT scanner and the treatment isocenter are then calculated. On the first day of treatment, the patient is first positioned to the initial reference marks and then shifted to the treatment isocenter using the calculated shifts. Initial reference marks are then removed and the treatment isocenter is marked on the patient. This process is illustrated in Figure 1.

 

With proper planning (from diagnostic workup), the initial marks can be placed very close to the center of target volume. With asymmetric jaws, the initial reference may be used as the isocenter, avoiding the shift on the treatment machine. However, often initial marks are not placed in an optimal location and shifts are necessary prior to treatment. These shifts can introduce uncertainties in patient treatments due to inexact shifting and patient re-marking, or occasionally due to incorrect shifts. Once there are incorrect shifts, exact corrections are very difficult to determine and some residual error inevitably remains. Additionally, this process can be inefficient as the patient has to be re-marked prior to treatment, occupying the treatment machine, or requiring separate treatment setup verification.

Figure 1 Shift Method: In this simulation method, initial patient marks are placed on the patient prior to the scan. During treatment planning, shifts (X, Y, Z) from the initial marks to the treatment isocenter are then determined. These shifts are applied prior to patient treatment and the patient is then re-marked.Figure 2 No-shift Method: In this simulation method, the treatment isocenter is marked while the patient is still in the CT simulator (Step 3). The patient goes home with marks which will be used for treatment. On the first treatment day there are no adjustments in patient setup, shifts, or remarking. This is exactly the same process as would be performed with a conventional simulator.
Figure 1
Figure 2
Shift Method: In this simulation method, initial patient marks are placed on the patient prior to the scan. During treatment planning, shifts (X, Y, Z) from the initial marks to the treatment isocenter are then determined. These shifts are applied prior to patient treatment and the patient is then re-marked.
No-shift Method: In this simulation method, the treatment isocenter is marked while the patient is still in the CT simulator (Step 3). The patient goes home with marks which will be used for treatment. On the first treatment day there are no adjustments in patient setup, shifts, or remarking. This is exactly the same process as would be performed with a conventional simulator.

 

No-shift method-for this method the patient is scanned and, while the patient is still on the CT scanner couch, images are reviewed and the treatment isocenter is determined based on CT images. The isocenter coordinates are then programmed into the movable lasers in the scanner room and the patient is marked accordingly. The patient then goes home with marks which will actually be used for treatment delivery. This is exactly how simulation with conventional simulator would be performed. This process is illustrated in Figure 2.

 

Determination of the isocenter can be performed by CT simulator staff, physician, or a dosimetrist, similar to what would be done with the conventional simulator. It is a common misconception that Tumor LOC is intended to be used by a physician. Tumor LOC at our institution is used primarily by therapists, and they are trained to place isocenters at optimal position for the majority of treatment sites according to physician's instructions. Physicians are typically only involved in isocenter placement for more complicated cases.

 

The No-shift method requires CT images to be reviewed using virtual simulation software in order to determine the isocenter that is to be marked on the patient. Historically, this software resided on a separate workstation. Such workstations are commercially available from several treatment planning software vendors. Images had to be transferred from the scanner and imported into a separate workstation. Image export, import, and processing to separate workstation takes almost ten minutes for most patients. In order to transfer the isocenter coordinates from virtual simulation software to the patient's skin accurately, the patient needs to remain still in the scanning position while the images are processed. Often patients are not able to remain still for this period of time. Since the Tumor LOC software resides directly on the CT scanner control console, the transfer of images is not required, and the isocenter location can be determined for most patients in two to four minutes. Therefore, with Tumor LOC software patients can be marked faster, reducing the possibility of patient movement between the scan and the actual marking. This is easily the greatest advantage of the Tumor LOC software compared to a separate workstation. It is true that this software eliminates an extra workstation in the scanner suite, but the ability to mark the patient in the shortest possible time is by far the greatest advantage. Tumor LOC software has tools that facilitate easy determination of optimal isocenter locations. For example, the software can locate the center of the patient on a particular CT image, so if the isocenter needs to be placed on that image, the point is located at the same depth in AP/ PA direction and also at the same depth in RT/LT lateral direction. This can all be performed without any contouring. There are other tools as well that facilitate efficient isocenter localization.

 

Another feature of Tumor LOC software and the CT Big Bore is absolute patient marking. Tumor LOC software can relate voxel positions in CT images directly to external laser coordinates inside the simulator room. If a point is selected on a CT image, the software can drive external lasers to coincide exactly with that voxel position for patient marking. The longitudinal CT couch position is the only step that requires manual movement. This is not the case with most standalone virtual simulation workstations. Standalone workstations generally cannot relate locations on CT images directly to the CT laser coordinates, and there has to be a shift relative to some reference marks which were placed on the patient prior to the scan. This is a similar process to the Shift method, except that the shift occurs on the CT scanner at time of simulation. As with other shift methods this leaves room for error. With absolute marking, if the CT couch has minimal sag, and lasers are accurately aligned with the image plane, patients can be marked directly based on locations determined on CT images. This feature significantly improves efficiency and accuracy of the No-shift method.

 

Tumor LOC is also a fully functional contouring environment. A physician can contour tumor volumes on the CT scanner control console, and these can be exported to the treatment planning workstation. In a well organized treatment planning process, the next time that the physician interacts with the CT study is at the time of final plan review. DICOM RT objects that are transferred from Tumor LOC to the treatment planning workstation are skin contours, any tumor volumes outlined by physician, isocenter location, and AP and lateral setup beams which were automatically generated by the software. When the CT study set is imported into the planning workstation it already has these structures built in and it is ready for further processing.

 

Tumor LOC software also has tools for the review and processing of respiratory correlated images (4D CT). The software allows contouring on 4D images and generation of surrogate images for creation of internal target volumes (ITVs). The full capability of Tumor LOC software with respect to 4D imaging is beyond the scope of this document.

 Anterior DRR with contour Right lateral DRR with contour
Anterior DRR with contour
Right lateral DRR with contour

 

Conclusions

Tumor LOC software enables the CT Big Bore simulator to function as a CT simulator rather than just as a CT scanner. The software can facilitate integration of the CT simulator into the clinical workflow to be used in similar fashion as the conventional simulator was used. The software can be used by therapists, physicians, or dosimetrists for accurate and efficient isocenter localization in the CT simulator room. Patients are transferred from the simulator to the treatment machine without any adjustments in position or skin marks, exactly the same as what was done with a conventional simulator. The CT images and structures created in Tumor LOC software can be transferred to any treatment planning system through DICOM connectivity.

 

To fully utilize functionality and efficiency of Tumor LOC software, it is imperative that staff allow appropriate time for training and evaluation of the necessary steps for integration of the CT Big Bore simulator into the clinical workflow. Once implemented, this scanner and software can improve efficiency and accuracy of patient treatments.



This content has been made possible by NetForum Community.
Share this on: Share your link in twitter Share your link in facebook Share your link on LinkedIn Print Rate this article: Log in to vote

 
Rating:
Votes:
1
Views:
5255
Added:
Apr 25, 2011

Rate this:
Log in to vote
 

White Paper
CT Big Bore
3D, CT simulation, Oncology, oncology body, oncology cervical spine, oncology extremity, oncology head, oncology lower extremity, oncology lumbar spine, oncology thoracic spine, oncology upper extremity, treatment planning
 

Clinical News
Best Practices
Case Studies
Publications and Abstracts
White Papers
Web seminars and Presentations
ExamCards
Protocols
Application Tips and FAQ
Training
Try an Application
Business News
Case Studies
White Papers
Web Seminars and Presentations
Utilization Services
Contributing Professionals
Contributing Institutions
Become a Contributor