Plato C. Lee

Patient dosimetry quality assurance program with a commercial diode system

PURPOSE To evaluate a commercial silicone diode dosimeter for a patient dosimetry quality assurance program. METHODS AND MATERIALS The diode dosimeter was calibrated against an ion chamber and percentage depth dose, linearity, anisotropy, virtual source position, and field size factor studies were performed. Correction factors for lack of full scatter medium in the diode entrance and exit dose measurements were acquired. Dosimetry equations were proposed for calculation of dose delivered at isocenter. Diode dose accuracy and reproducibility were tested on phantom and on four patients. A patient dosimetry quality assurance program based on diode measured dose was instituted and patient dose data were collected. RESULTS Diode measured percentage depth dose and field factors agreed to within 3% with those measured with an ion chamber. The diode exhibited less than 1.7% angular dose anisotropy and less than 0.5% nonlinearity up to 4 Gy. Diode dose measurements in phantom showed that the calculated doses differed from the prescribed dose by less than 1.5%; the diode exhibited a daily dose reproducibility of better than 0.2%. On four selected patients, the measured dose reproducibility was 1.5%; the average calculated doses were all within +/- 7% of the prescribed doses. For 33 of 40 patients treated with a 6 MV beam, measured doses were within +/- 7% of the prescribed doses. For 58 of 63 patients treated with an 18 MV beam, measured doses were within +/- 7% of the prescribed doses. For 11 out of 12 patients, a second repeat measurements yielded doses within +/- 7% of the prescribed doses. CONCLUSIONS The proposed diode-based patient dosimetry quality assurance program with dose tolerance at +/- 7% is simple and feasible. It is capable of detecting certain serious treatment errors such as incorrect daily dose greater than 7%, incorrect wedge use, incorrect photon energy and patient setup errors involving some incorrect source-to-surface-distance vs. source-to-axis-distance treatments.

Monte Carlo simulations of the differential beam hardening effect of a flattening filter on a therapeutic x‐ray beam

A quantitative study of the differential beam hardening effect of the flattening filter on the 6-MV beam of Clinac 2100C has been conducted with Monte Carlo simulations using EGS4 code. The fluence-weighted photon energy of the unfiltered beam decreases from 1.35 MeV at central axis (CAX) to 1.22 MeV at an off-axis distance (OAD) of 20.0 cm. Compared to the unfiltered beam, the fluence-weighted photon energy of the filtered beam increases to 1.93 MeV at CAX and to 1.36 MeV at an OAD f 20.0 cm, respectively. The beam hardening effect was found to be 2.1 times higher at CAX than at an OAD of 20.0 cm. With the differential filtration of the flattening filter, the photon energy fluence reduced to 44% and 78% at CAX and an OAD of 20.0 cm respectively, resulting in the energy fluence of the filtered beam being flat from CAX to an OAD of 20.0 cm. The differential transmission ratios between the high energy and low energy photons decrease as the OAD increases. The percentage depth doses (PDDs) at field size of 10.0 cm x 10.0 cm for both the filtered and unfiltered 6-MV beams at CAX and at an OAD of 15.0 cm were calculated with a Monte Carlo technique based on the simulated spectra and fluence. The calculated PDDs were found to be consistent with the measured data for the filtered beam at CAX and an OAD of 15.0 cm. The beam quality (BQ) of the filtered beam at CAX is also higher than that of the same beam at an OAD of 15.0 cm. All the above results quantitatively demonstrate the differential beam hardening effects of a flattening filter on a therapeutic x-ray beam.

Consistent collimator overlaps in field matching with computer-controlled x-ray collimators

Consistent collimator overlaps in field matching with computer-controlled asymmetric x-ray collimators have been observed and investigated. The results indicate that, although this consistent overlapping is small and within manufacturers +/- 1 mm discrepancy specification, it can produce more than 10% cold spots at matchlines if unheeded. This undesirable dose inhomogeneity, however, can be improved to within +/- 3% with a simple and practical solution of artificially offsetting the collimators by exactly the same amount (e.g., 1 mm) they overlap.

Seed loss through the urinary tract after prostate brachytherapy: examining the role of cystoscopy and urine straining post implant

This study describes one institutions experience with seed retrieval through the urinary tract and makes recommendations for cystoscopy and urine straining post prostate brachytherapy (PB). 1794 patients from two separate cohorts covering different time periods (early versus late) were analyzed. All patients were preplanned with a modified peripheral loading technique and implanted with preloaded needles (125I or 103Pd) under ultrasound guidance. A catheter was used to delineate the urethra during the volume study but was not used during the implant. All patients underwent post implant cystoscopy. All patients were instructed to strain their urine for seven days post implant and return any seeds to our center. In our experience, seed loss through the urinary tract is a common event after PB, occurring in 29.7% of patients and was more common in patients from the early cohort, those implanted with 125I seeds or those patients with prior transurethral resection of the prostate. Average seed loss per case, however, represents only 0.58% of total activity. We continue to recommend routine post implant cystoscopy for seed retrieval and periprocedural management. We no longer recommend that patients strain their urine at home after documenting a low rate of seed loss after discharge.

TECHNIQUE CHARTS FOR KODAK'S NEW FILM-SCREEN SYSTEMS FOR PORTAL LOCALIZATION

In July 1996, Kodak released new film-screen systems with enhanced contrast (EC) for portal localization with megavoltage therapeutic beams. This study presents the generation of general-purpose technique charts for Kodaks new film-screen combinations: the Enhanced-Contrast localization (EC-L) film in EC-L cassette and the EC-L film in fast ECL (fECL) cassette for use with cobalt-60, 6 MV, 10 MV, and 18 MV beams. These technique charts were based on the assumption that a film with an optical density (OD) of 1.8 provides the best viewing density. The doses to produce such as OD, Dexp, were obtained from the H & D curves and were 1.5, 1.6, 1.7, and 1.8 cGy for cobalt-60, 6 MV, 10 MV, and 18 MV beams, respectively, with the EC-L-film + EC-L-cassette combination. The corresponding values were 1.3, 1.3, 1.3, and 1.4 cGy, respectively, for the above four beams with the EC-L-film + fEC-L-cassette combination. The dose to the film is assumed to be proportional to the calibrated dose rate (D0), field size factor (FSF), inverse square factor relative to 100.0 cm (INV), and the transmission factor through the patient, which is equal to e-uT, where u is the broad beam attenuation coefficient and T is the patient thickness. With the above assumptions, the exposure time or monitor unit, t, is then calculated from the following equation: t = Dexp/(D0*FSF*INV*e-uT). For an average port size of 15 x 15 cm, the attenuation coefficients were obtained from the fitting of TAR (cobalt-60) or TMR (6 MV, 10 MV, and 18 MV) as a function of depth from 10 to 30 cm and were 0.0564 cm-1, 0.03714 cm-1, and 0.02271 cm-1 for cobalt-60, 6 MV, 10 MV, and 18 MV beams, respectively. The FSF were explicitly obtained from the clinical physics data books and were 1.028, 1.032, 1.036, and 1.053 for cobalt-60, 6 MV, 10 MV, and 18 MV, respectively. For cobalt-60 beam, the D0 was assumed to be 100.0 cGy/min. For the 6 MV, 10 MV, and 18 MV beams, the D0 in cGy per monitor unit is 1.030, 1.051, and 1.061, respectively. Technique charts were then generated as a function of patient thickness from 10 to 45 cm for filming distance from 110 to 140 cm for all four beams. These technique charts can be easily customized to portal localization practices in a radiation therapy department.

Verification of a simple point dose calculation method for a dual energy linear accelerator with asymmetric jaws

Certain fundamental dosimetrical parameters involving the applications of asymmetric jaws were investigated. The nominal accelerating potentials (NAPs) were found to decrease from 5.1 to 4.2 and from 18.0 to 13.4 for the 6 and 18 MV beams, respectively, as the off-axis distance (OAD) increases from 0.0 to 15.0 cm. The relative beam intensity increases from 1.00 to 1.07 at OAD of 15.0 cm for the 6 MV beam, and to 1.02 at OAD of 7.0 cm for the 18 MV beam. The percentage depth doses (PDDs) for half-blocked fields of 4 x 4 cm, 10 x 10 cm and 20 x 20 cm were found to deviate from those of corresponding symmetric fields by less than 2% down to the depth of 35.0 cm. The field size factor (FSF) for the asymmetric field from 4 x 4 cm to 20 x 20 cm deviates less than 1.0% from those of the corresponding symmetric fields. The equivalent square concept was found to be applicable to asymmetric fields within 1% error if the jaw exchange effect is taken into consideration. The measured point doses for half-blocked fields of 4 x 4 cm, 10 x 10 cm and 20 x 20 cm for both 6 and 18 MV were within 3% of the calculated dose based on a published dose calculation method which employs symmetric field beam parameters, such as field size factor (FSF), percentage depth dose (PDD), and off-axis correction factors (OAFs). The efficacy of this point dose calculation method is discussed.

Characteristics of a spoiled 6‐MV beam from a dual‐energy linear accelerator

A beam spoiler for a 6-MV x-ray beam was designed and spoiled beam dosimetry performed. The spoiled beam quality was similar to that of the unspoiled beam. The percentage depth doses at 5 mm were at least 90% for all field sizes except 4 x 4 cm, while the surface dose at 0 mm varied from 22.5% for a 4 x 4-cm field to 85.8% for a 22 x 22-cm field. The spoiled beam showed degraded beam flatness for small field size at superficial depths; the beam flatness at 6 mm for a 4 x 4-cm field was +/- 9.0%. The beam flatness improved as the field size increased, e.g., at same depth, the beam flatness was +/- 2.5% for a 22 x 22-cm field. The penumbra width (90%-10%) of the spoiled beam was greater than that of the unspoiled beam for all field sizes and at all depths; e.g., at 0-mm depth, for a 22 x 22-cm field, the penumbra width was 3.8 cm for the unspoiled beam and 7.2 cm for the spoiled beam. Beyond dmax the difference in the penumbra widths between the spoiled and unspoiled beam was about 3-4 mm for all field sizes and at all depths. The peripheral dose was larger for the spoiled beam.

Beam-hardening effects of wedges on a spoiled 6 MV beam

The beam-hardening effects of the wedges on a 6-MV spoiled beam has been studied. The beam quality of all the wedged beams was found to be the same as the open beam. The dmax also stayed unchanged for all the wedges at all field sizes. The relative wedge factors were found to reflect the beam-hardening effect, which is a function of the wedge angles, depths, and field sizes. For the 15 degrees and 30 degrees wedges, the relative wedge factor at depths less than 15 cm were found to deviate less than +2% for all fields, while those for the 45 degrees and 60 degrees wedges for the same depth range from +3% to +4%. The surface doses were found to decrease with the wedged fields. For 15 degrees, 30 degrees, and 45 degrees wedges, the decreases were found to be from 0 to -2.0%. For 60 degrees wedges, the largest deviation was found to be -2.5% for a field size of 10 cm x 10 cm at a depth of 2 mm. The wedge factors at dmax were found to depend slightly on the field sizes. The use of an averaged wedge factor for each individual wedge was found to produce less than +/- 1.2% of error for all field sizes.