Multimodality image registration quality assurance for conformal three-dimensional treatment planning.

PURPOSE: We present a quality assurance methodology to determine the accuracy of multimodality image registration and fusion for the purpose of conformal three-dimensional and intensity-modulated radiation therapy treatment planning. Registration and fusion accuracy between any combination of computed tomography (CT), magnetic resonance (MR), and positron emission computed tomography (PET) imaging studies can be evaluated. METHODS AND MATERIALS: A commercial anthropomorphic head phantom filled with water and containing CT, MR, and PET visible targets was modified to evaluate the accuracy of multimodality image registration and fusion software. For MR and PET imaging, the water inside the phantom was doped with CuNO(3) and 18F-fluorodeoxyglucose (18F-FDG), respectively. Targets consisting of plastic spheres and pins were distributed throughout the cranium section of the phantom. Each target sphere had a conical-shaped bore with its apex at the center of the sphere. The pins had a conical extension or indentation at the free end. The contours of the spheres, sphere centers, and pin tips were used as anatomic landmark models for image registration, which was performed using affine coordinate-transformation tools provided in a commercial multimodality image registration/fusion software package. Four sets of phantom image studies were obtained: primary CT, secondary CT with different phantom immobilization, MR, and PET study. A novel CT, MR, and PET external fiducial marking system was also tested. RESULTS: The registration of CT/CT, CT/MR, and CT/PET images allowed correlation of anatomic landmarks to within 2 mm, verifying the accuracy of the registration software and spatial fidelity of the four multimodality image sets. CONCLUSIONS: This straightforward phantom-based quality assurance of the image registration and fusion process can be used in a routine clinical setting or for providing a working image set for development of the image registration and fusion process and new software.

A novel approach to overcome hypoxic tumor resistance: Cu-ATSM-guided intensity-modulated radiation therapy.

PURPOSE: Locoregional tumor control for locally advanced cancers with radiation therapy has been unsatisfactory. This is in part associated with the phenomenon of tumor hypoxia. Assessing hypoxia in human tumors has been difficult due to the lack of clinically noninvasive and reproducible methods. A recently developed positron emission tomography (PET) imaging-based hypoxia measurement technique which employs a Cu(II)-diacetyl-bis(N(4)-methylthiosemicarbazone) (Cu-ATSM) tracer is of great interest. Oxygen electrode measurements in animal experiments have demonstrated a strong correlation between low tumor pO(2) and excess (60)Cu-ATSM accumulation. Intensity-modulated radiation therapy (IMRT) allows selective targeting of tumor and sparing of normal tissues. In this study, we examined the feasibility of combining these novel technologies to develop hypoxia imaging (Cu-ATSM)-guided IMRT, which may potentially deliver higher dose of radiation to the hypoxic tumor subvolume to overcome inherent hypoxia-induced radioresistance without compromising normal tissue sparing. METHODS AND MATERIALS: A custom-designed anthropomorphic head phantom containing computed tomography (CT) and positron emitting tomography (PET) visible targets consisting of plastic balls and rods distributed throughout the "cranium" was fabricated to assess the spatial accuracy of target volume mapping after multimodality image coregistration. For head-and-neck cancer patients, a CT and PET imaging fiducial marker coregistration system was integrated into the thermoplastic immobilization head mask with four CT and PET compatible markers to assist image fusion on a Voxel-Q treatment-planning computer. This system was implemented on head-and-neck cancer patients, and the gross tumor volume (GTV) was delineated based on physical and radiologic findings. Within GTV, regions with a (60)Cu-ATSM uptake twice that of contralateral normal neck muscle were operationally designated as ATSM-avid or hypoxic tumor volume (hGTV) for this feasibility study. These target volumes along with other normal organs contours were defined and transferred to an inverse planning computer (Corvus, NOMOS) to create a hypoxia imaging-guided IMRT treatment plan. RESULTS: A study of the accuracy of target volume mapping showed that the spatial fidelity and imaging distortion after CT and PET image coregistration and fusion were within 2 mm in phantom study. Using fiducial markers to assist CT/PET imaging fusion in patients with carcinoma of the head-and-neck area, a heterogeneous distribution of (60)Cu-ATSM within the GTV illustrated the success of (60)Cu-ATSM PET to select an ATSM-avid or hypoxic tumor subvolume (hGTV). We further demonstrated the feasibility of Cu-ATSM-guided IMRT by showing an example in which radiation dose to the hGTV could be escalated without compromising normal tissue (parotid glands and spinal cord) sparing. The plan delivers 80 Gy in 35 fractions to the ATSM-avid tumor subvolume and the GTV simultaneously receives 70 Gy in 35 fractions while more than one-half of the parotid glands are spared to less than 30 Gy. CONCLUSION: We demonstrated the feasibility of a novel Cu-ATSM-guided IMRT approach through coregistering hypoxia (60)Cu-ATSM PET to the corresponding CT images for IMRT planning. Future investigation is needed to establish a clinical-pathologic correlation between (60)Cu-ATSM retention and radiation curability, to understand tumor re-oxygenation kinetics, and tumor target uncertainty during a course of radiation therapy before implementing this therapeutic approach to patients with locally advanced tumor.

Initial experience with quality assurance of multi-institutional 3D radiotherapy clinical trials. A brief report.

In 1992, a 3D Quality Assurance (3D QA) Center was established at the Mallinckrodt Institute of Radiology under the auspices of the Radiation Therapy Oncology Group (RTOG). The role of the 3D QA Center is to provide quality assurance reviews of external beam treatment planning and verification (TPV) information for patients enrolled in multi-institutional 3D radiotherapy treatment protocols. Computer hardware and software components have been implemented which allow participating institutions to submit (via either the Internet or magnetic tape) common format 3D TPV data for QA review including: volumetric CT image data, normal structure, tumor and target volume contours, digitally reconstructed radiographs or simulator (prescription) and portal radiographs, beam geometry, dose distributions, fractionation information, and dose-volume histograms. Prior to enrolling patients on a 3D radiotherapy treatment protocol, each participating institution is required to complete a 3D Facility Questionnaire documenting their 3D treatment planning capability. In addition, the successful completion of a protocol "dry run" test is required to demonstrate the participating institution's ability to submit a protocol complaint digital data set to the 3D QA Center prior to placing patients on the 3D CRT study. Two site specific (prostate and lung) phase I/II 3D dose escalation trials are currently accruing patients. The QA center reviews at a minimum the first 5 cases from each participating institution and spot checks subsequent submissions. For each case review the following parameters are evaluated: 1. data exchange compliance, 2. CT data quality, 3. target volume contours, 4. normal structure contours, 5. field placement, 6. field shape, 7. dose prescription, 8. dose uniformity, and 9. dose conformity. By April 1997, over 300 protocol patient TPV data sets have been submitted and reviewed by the 3D QA Center.

A critical evaluation of the planning target volume for 3-D conformal radiotherapy of prostate cancer.

PURPOSE: To determine an adequate planning target volume (PTV) margin for three-dimensional conformal radiotherapy (3D CRT) of prostate cancer, the uncertainties in the internal positions of the prostate and seminal vesicles (SV) and in the treatment setups were measured. METHODS AND MATERIALS: Weekly computed tomography (CT) scans of the pelvis (n=51) and daily electronic portal images (n=1630) were reviewed for eight patients who received seven-field 3D CRT for prostate cancer. The CT scans were registered in three dimensions to the original planning CT scan using commercially available software to measure the center-of volume (COV) motion of the prostate and SV. The daily portal images were registered to the corresponding simulation films to measure the setup displacements. The standard deviation (SD) of the internal organ motions was added to the SD of the setups in quadrature to determine the total uncertainty. Positive directions were left, anterior, and superior. Rotations necessary to register the CT scans and portal images were minimal and not further analyzed. RESULTS: The mean motion for the COV of the prostate+/-the SD was 0+/-0.9 mm in the left-right (LR), 0.5+/-2.6 mm in the anterior-posterior (AP), and 1.5+/-3.9 mm in the superior-inferior (SI) directions. The mean motion for the COV of the SV+/-the SD was 0.3+/-1.7 mm in the LR, 0.7+/-3.8 mm in the AP, and 0.9+/-3.5 mm in the SI directions. For all patients the mean isocenter displacement+/-the SD was 0+/-3.1 mm in the LR, 1.4+/-3.0 mm in the AP, and -0.4+/-2.1 mm in the SI directions. The total uncertainty for the prostate was 3.2 mm, 4.0 mm, and 4.4 mm in the LR, AP, and SI directions, respectively. For the SV, the total uncertainty was 3.5, 4.8, and 4.1 mm in the LR, AP, and SI directions, respectively. CONCLUSIONS: PTV margins of 10 to 16 mm are required to encompass all (99%) possible positions of the prostate or SV during 3D CRT. PTV margins of 7 to 11 mm will encompass the measured uncertainties with a 95% probability. PTV margins of 5 mm may not adequately cover the intended volume.

Quantitative dosimetric verification of an IMRT planning and delivery system.

BACKGROUND AND PURPOSE: The accuracy of dose calculation and delivery of a commercial serial tomotherapy treatment planning and delivery system (Peacock. NOMOS Corporation) was experimentally determined. MATERIALS AND METHODS: External beam fluence distributions were optimized and delivered to test treatment plan target volumes, including three with cylindrical targets with diameters ranging from 2.0 to 6.2 cm and lengths of 0.9 through 4.8 cm, one using three cylindrical targets and two using C-shaped targets surrounding a critical structure, each with different dose distribution optimization criteria. Computer overlays of film-measured and calculated planar dose distributions were used to assess the dose calculation and delivery spatial accuracy. A 0.125 cm3 ionization chamber was used to conduct absolute point dosimetry verification. Thermoluminescent dosimetry chips, a small-volume ionization chamber and radiochromic film were used as independent checks of the ion chamber measurements. RESULTS: Spatial localization accuracy was found to be better than +/-2.0 mm in the transverse axes (with one exception of 3.0 mm) and +/-1.5 mm in the longitudinal axis. Dosimetric verification using single slice delivery versions of the plans showed that the relative dose distribution was accurate to +/-2% within and outside the target volumes (in high dose and low dose gradient regions) with a mean and standard deviation for all points of -0.05% and 1.1%, respectively. The absolute dose per monitor unit was found to vary by +/-3.5% of the mean value due to the lack of consideration for leakage radiation and the limited scattered radiation integration in the dose calculation algorithm. To deliver the prescribed dose, adjustment of the monitor units by the measured ratio would be required. CONCLUSIONS: The treatment planning and delivery system offered suitably accurate spatial registration and dose delivery of serial tomotherapy generated dose distributions. The quantitative dose comparisons were made as far as possible from abutment regions and examination of the dosimetry of these regions will also be important. Because of the variability in the dose per monitor unit and the complex nature of the calculation and delivery of serial tomotherapy, patient-specific quality assurance procedures will include a measurement of the delivered target dose.

Real-time 3D dose calculation and display: a tool for plan optimization.

PURPOSE: Both human and computer optimization of treatment plans have advantages; humans are much better at global pattern recognition, and computers are much better at detailed calculations. A major impediment to human optimization of treatment plans by manipulation of beam parameters is the long time required for feedback to the operator on the effectiveness of a change in beam parameters. Our goal was to create a real-time dose calculation and display system that provides the planner with immediate (fraction of a second) feedback with displays of three-dimensional (3D) isodose surfaces, digitally reconstructed radiographs (DRRs), dose-volume histograms, and/or a figure of merit (FOM) (i.e., a single value plan score function). This will allow the experienced treatment planner to optimize a plan by adjusting beam parameters based on a direct indication of plan effectiveness, the FOM value, and to use 3D display of target, critical organs, DRRs, and isodose contours to guide changes aimed at improving the FOM value. METHODS AND MATERIALS: We use computer platforms that contain easily utilized parallel processors and very tight coupling between calculation and display. We ported code running on a network of two workstations and an array of transputers to a single multiprocessor workstation. Our current high-performance graphics workstation contains four 150-MHz processors that can be readily used in a shared-memory multithreaded calculation. RESULTS: When a 10 x 10-cm beam is moved, using an 8-mm dose grid, the full 3D dose matrix is recalculated using a Bentley-Milan-type dose calculation algorithm, and the 3D dose surface display is then updated, all in < 0.1s. A 64 x 64-pixel DRR calculation can be performed in < 0.1 s. Other features, such as automated aperture calculation, are still required to make real-time feedback practical for clinical use. CONCLUSION: We demonstrate that real-time plan optimization using general purpose multiprocessor workstations is a practical goal. Parallel processing technology provides this capability for 3D planning systems, and when combined with objective plan ranking algorithms should prove effective for optimizing 3D conformal radiation therapy. Compared to our earlier transputer work, multiprocessor workstations are more easily programmed, making software development costs more reasonable compared with uniprocessor development costs. How the dose calculation is partitioned into parallel tasks on a multiprocessor work station can make a significant difference in performance. Shared-memory multiprocessor workstations are our first choice for future work, because they require minimum programming effort and continue to be driven to higher performance by competition in the workstation arena.

Prospective clinical evaluation of an electronic portal imaging device.

PURPOSE: To determine whether the clinical implementation of an electronic portal imaging device can improve the precision of daily external beam radiotherapy. METHODS AND MATERIALS: In 1991, an electronic portal imaging device was installed on a dual energy linear accelerator in our clinic. After training the radiotherapy technologists in the acquisition and evaluation of portal images, we performed a randomized study to determine whether online observation, interruption, and intervention would result in more precise daily setup. The patients were randomized to one of two groups: those whose treatments were actively monitored by the radiotherapy technologists and those that were imaged but not monitored. The treating technologists were instructed to correct the following treatment errors: (a) field placement error (FPE) > 1 cm; (b) incorrect block; (c) incorrect collimator setting; (d) absent customized block. Time of treatment delivery was recorded by our patient tracking and billing computers and compared to a matched set of patients not participating in the study. After the patients radiation therapy course was completed, an offline analysis of the patient setup error was planned. RESULTS: Thirty-two patients were treated to 34 anatomical sites in this study. In 893 treatment sessions, 1,873 fields were treated (1,089 fields monitored and 794 fields unmonitored). Ninety percent of the treated fields had at least one image stored for offline analysis. Eighty-seven percent of these images were analyzed offline. Of the 1,011 fields imaged in the monitored arm, only 14 (1.4%) had an intervention recorded by the technologist. Despite infrequent online intervention, offline analysis demonstrated that the incidence of FPE > 10 mm in the monitored and unmonitored groups was 56 out of 881 (6.1%) and 95 out of 595 (11.2%), respectively; p < 0.01. A significant reduction in the incidence of FPE > 10 mm was confined to the pelvic fields. The time to treat patients in this study was 10.78 min (monitored) and 10.10 min (unmonitored). Features that were identified that prevented the technologists from recognizing more errors online include poor image quality inherent to the portal imaging device used in this study, artifacts on the portal images related to table supports, and small field size lacking sufficient anatomical detail to detect FPEs. Furthermore, tools to objectively evaluate a portal image for the presence of field placement error were lacking. These include magnification factor corrections between the simulation of portal image, online measurement tools, image enhancement tools, and image registration algorithms. CONCLUSION: The use of an electronic portal imaging device in our clinic has been implemented without a significant increase in patient treatment time. Online intervention and correction of patient positioning occurred rarely, despite FPEs of > 10 mm being present in more than 10% of the treated fields. A significant reduction in FPEs exceeding 10 mm was made in the group of patients receiving pelvic radiotherapy. It is likely that this improvement was made secondarily to a decrease in systematic error and not because of online interventions. More significant improvements in portal image quality and the availability of online image registration tools are required before substantial improvements can be made in patient positioning with online portal imaging.

An analysis of intratreatment and intertreatment displacements in pelvic radiotherapy using electronic portal imaging.

PURPOSE: To evaluate the relative frequency and magnitude of intratreatment and intertreatment displacements in the patient positioning for pelvic radiotherapy using electronic portal imaging. METHODS AND MATERIALS: Five hundred ninety-four electronic portal images of seven patients treated with a four-field pelvic technique were evaluated. All patients were treated prone without an immobilization device. Two fields were treated per day, from which an average of two electronic portal images were obtained for each field. No treatment was interrupted or adjusted on the basis of these images. Each image was aligned to the corresponding simulation film to measure the displacements in the mediolateral, craniocaudal, and anteroposterior directions relative to the simulated center. The intertreatment displacement was the displacement measured from the initial image for each daily treated field. For each daily treated field the intratreatment displacement was calculated by subtracting the displacement measured on the initial image from the displacement measured on the final image. RESULTS: The frequency of the intertreatment displacements exceeding 10 mm was 3%, 16%, and 23% for the mediolateral, craniocaudal, and anteroposterior translations, respectively. There were no intratreatment displacements exceeding 10mm (p < 0.001). The frequency of intertreatment displacements exceeded 5 mm was 40, 52, and 51% for the mediolateral, craniocaudal, and anteroposterior translations, respectively; whereas, the frequency of intratreatment displacements exceeding 5 mm was 1, 5, and 7% for the same translations, respectively (p < 0.001). The standard deviation of the intertreatment displacements was at least three times as great as the standard deviation of the intratreatment displacements for all translations. These deviations were greater than the precision limit of the measurement technique, which is approximately 1mm. Each patient had one direction where systematic error predominated in intertreatment positioning. Random error predominated for intratreatment positioning and for the other two directions in intertreatment positioning. CONCLUSIONS: During a course of pelvic radiotherapy, the frequency of intertreatment displacements exceeding 5 and 10 mm is significantly greater than the frequency of intratreatment displacements of these magnitudes. Errors in intertreatment positioning are predominantly systematic in one direction for each patient, whereas intratreatment error is predominantly random. Because patients do not move considerably during the daily treatment of a pelvic field, a single electronic portal image per daily field may be considered representative of the treated position.

The Electronic View Box: a software tool for radiation therapy treatment verification.

PURPOSE: We have developed a software tool for interactively verifying treatment plan implementation. The Electronic View Box (EVB) tool copies the paradigm of current practice but does so electronically. A portal image (online portal image or digitized port film) is displayed side by side with a prescription image (digitized simulator film or digitally reconstructed radiograph). The user can measure distances between features in prescription and portal images and "write" on the display, either to approve the image or to indicate required corrective actions. The EVB tool also provides several features not available in conventional verification practice using a light box. METHODS AND MATERIALS: The EVB tool has been written in ANSI C using the X window system. The tool makes use of the Virtual Machine Platform and Foundation Library specifications of the NCI-sponsored Radiation Therapy Planning Tools Collaborative Working Group for portability into an arbitrary treatment planning system that conforms to these specifications. The present EVB tool is based on an earlier Verification Image Review tool, but with a substantial redesign of the user interface. A graphical user interface prototyping system was used in iteratively refining the tool layout to allow rapid modifications of the interface in response to user comments. RESULTS: Features of the EVB tool include 1) hierarchical selection of digital portal images based on physician name, patient name, and field identifier; 2) side-by-side presentation of prescription and portal images at equal magnification and orientation, and with independent grayscale controls; 3) "trace" facility for outlining anatomical structures; 4) "ruler" facility for measuring distances; 5) zoomed display of corresponding regions in both images; 6) image contrast enhancement; and 7) communication of portal image evaluation results (approval, block modification, repeat image acquisition, etc.). CONCLUSION: The EVB tool facilitates the rapid comparison of prescription and portal images and permits electronic communication of corrections in port shape and positioning.

Is weekly port filming adequate for verifying patient position in modern radiation therapy?

PURPOSE: The objective of this study is to use daily electronic portal imaging to evaluate weekly port filming in detecting patient set-up position. METHODS AND MATERIALS: A computer-based portal alignment method was used to quantify the field displacements on 191 digitized weekly port films and 848 daily electronic portal images in 21 radiation therapy patients. An electronic portal image data set as a control for actual daily treatment position was used to evaluate weekly port films with respect to same-day field displacement, rate of field placement error detection, and prediction of subsequent daily field displacements. RESULTS: The field displacements measured on a port film frequently deviated from the corresponding field displacements on the electronic portal image obtained in the same treatment set-up. A linear regression analysis showed that the curves fitted to the same-day field displacements had slopes that differed significantly from unity (p < 0.001). Overall, the respective frequencies of field placement error, beyond clinical tolerance limits of 5, 7, and 10 mm (corresponding to head and neck, thoracic, and pelvic sites) for port filming and electronic portal imaging were 11% and 14% (p = 0.4) in the X-direction (lateral or anteroposterior) and 24% and 13% (p = .0001) in the Y-direction (caphalad-caudad). When the data were broken down by anatomical region, this discrepancy was found to be mainly due to the differences in the thorax, and head and neck image data sets. For thoracic fields, error in Y-shifts was 28% by port filming, but only 9% by portal imaging (p = 0.01). In the head and neck region, 18% of the port films exceeded tolerance, whereas only 6% of the electronic portal images did (p = 0.0001). Field displacements on the treatment set-ups between the acquisition of port films were not predicted by those films. CONCLUSION: There are discrepancies between the field displacements and field placement errors detected by weekly port films and daily electronic portal images. This study suggests that improved methods of treatment verification may be necessary in modern radiation therapy.

An evaluation of two methods of anatomical alignment of radiotherapy portal images.

PURPOSE: Two techniques have been developed at our institution to allow anatomical registration of digitized portal images to a simulation film. Accuracy of the portal image alignment methods is tested and single intrauser and multiple interuser variation is examined using each technique. METHODS AND MATERIALS: Method one requires the identification of anatomical fiducial points on a simulation image and its corresponding portal image. The parameters required to align the corresponding points are calculated by a least squares fit algorithm. Method two uses an anatomical template generated from the simulation image and superimposing it upon a portal image. The template is then adjusted by a computer mouse to obtain the best subjective anatomical fit on the portal image. Megavoltage portal images of a skull phantom with various known shifts and eight clinical image files were aligned by each method. Each data set was aligned several times by both a single user and multiple users. RESULTS: Alignment of the anatomical phantom portal images demonstrates an accuracy of less than 0.8 +/- 0.9 mm and 0.7 +/- 1.0 degrees with either method. As out of plane rotation increased from 0 to 5 degrees, simulating out of plane malpositioning, alignment orthogonal to the plane of rotation worsened to 1.5 +/- 1.1 mm with the point method and 2.4 +/- 1.6 mm with the template method. Alignment parallel to the axis of the gantry rotation was insensitive to this change and remained constant as did the rotational alignment parameters. For the clinical image files the magnitude of variation for a single user is typically less than +/- 1 mm or +/- 1 degree. The magnitude of variation of alignment increased when multiple users aligned the same image files. The variation was dependent upon anatomical site and to a lesser degree the method of alignment used. The root mean square deviation of translational shifts range from +/- 0.68 mm when using the template method in the pelvis to as high as +/- 2.94 mm with the template method to align abdominal portal images. In the thorax and pelvis translational alignments along the horizontal axis were more precise than along the vertical axis. Multiple user variability was in part due to poor image quality, user experience, non rigidity of the anatomical features, and the difficulty in locating an exact point on a continuous anatomical structure. CONCLUSION: In well controlled phantom studies both the fiducial point and template method provide similar and adequate results. The phantom studies show that alignment error and variance increase with distortion in anatomical features secondary to out of plane rotations. In clinical situations intrauser variation is small, however, multiple interuser variation is larger. The magnitude of variation is dependent upon the anatomical site aligned.

Effects of spike discharge history on discharge probability and latency in frog basilar papilla units.

Gaumond et al. [(1982) J. Neurophysiol. 48, 856-873] showed in the cat that a multiplicative-intensity model can generally account quite well for reduction of the probability of an auditory-nerve spike by another spike preceding it by 4 to 25 ms, and that for smaller separations there is also an increased latency of the following spike. Bosch [(1990) D. Sc. Dissertation, Washington University, St. Louis, MO] made important improvements in experimental design and estimation techniques for studying these effects, and confirmed their presence in the gerbil. However, direct application of these methods to the frog does not yield reliable estimates. A clearer separation of discharge probability and latency effects in frog basilar papilla units is provided by the paired-click paradigm used in this study, which is applicable to low-spontaneous-rate units that generally respond to click stimuli with zero or one spike within a short interval following the click. The results confirm the existence in the frog of both spike-probability and spike-latency effects that are qualitatively similar to those found in mammals, although the absolute refractory time is much longer in frog, and the relative refractory time usually shorter. The paired-click paradigm also reveals a stimulus-history effect at stimulus levels which are near threshold: when there is no response to the first click, responses to the second click occur with increased probability and reduced latency.