Volume 3, Issue 1, January 1976
Index of content:
3(1976); http://dx.doi.org/10.1118/1.594270View Description Hide Description
An overview of Bayesian statistical decision theory is presented in the tutorial spirit. A section on fundamental principles is followed by selected applications of the Bayesian approach to parameter estimation, pattern recognition,image processing, computer‐aided medical diagnosis, optimal diagnostic test selection, and radiotherapytreatment planning.
3(1976); http://dx.doi.org/10.1118/1.594223View Description Hide Description
A computer program for the generation of high‐energy x‐ray dose distributions has been developed. The program takes account of the fact that the x‐ray beam profile, as affected by penumbra, wedges, beam blocks, or flattening filters, is one of the prime determinants of the dose distribution. Doses from primary and scattered radiation are calculated separately by using published tissue‐air ratio (TAR) and scatter‐air ratio (SAR) data. The dose from scattered radiation is determined by dividing the x‐ray field into a series of scatter strips and summing the contribution from each. Wedges or beam blocks are taken into account simply by entering their actual coordinates and linear attenuation coefficients. The program may be used to design wedges, beam blocks, or flattening filters prior to fabrication. Complete libraries of dose distributions for 60Co, 2‐, 4‐, and 6‐MV x rays have been generated.
3(1976); http://dx.doi.org/10.1118/1.594224View Description Hide Description
A test cassette employing a modification of the Ardran–Crooks attenuation technique for measuring peak tube potential (and half‐value layer) is described. The technique employs a radiographic intensifying screen which is exposed to a hardened x‐ray beam. Copper penetrameters are used to determine the attenuator thickness which reduces the light output from the screen to that transmitted by an optical attenuator over an adjacent area of the screen. The test cassette was calibrated by using x‐ray machines whose high voltages were measured electrically. The test cassette has a measurement precision of ±1 kVp and an accuracy of ±3 kVp or better in the range 50–130 kVp for x‐ray generators with the same voltage waveform.
3(1976); http://dx.doi.org/10.1118/1.594225View Description Hide Description
Beam flattening by the use of polyethylene filters has been developed for the 50‐MeV d→Be fast‐neutron therapy beam at the Texas A&M Variable‐Energy Cyclotron (TAMVEC) as a result of the need for a more uniform dose distribution at depth within the patient. A computer algorithm has been developed that allows the use of a modified decrement line method to calculate dose distributions; standard decrement line methods do not apply because of off‐axis peaking. The dose distributions for measured flattened beams are transformed into distributions that are physically equivalent to an unflattened distribution. In the transformed space, standard decrement line theory yields a distribution for any field size which, by applying the inverse transformation, generates the flattened dose distribution, including the off‐axis peaking. A semiempirical model has been constructed that allows the calculation of dose distributions for wedged beams from open‐beam data.
3(1976); http://dx.doi.org/10.1118/1.594226View Description Hide Description
Some of the physical parameters necessary in electron beam treatment planning are investigated for 5‐ to 45‐MeV electrons extracted from a Brown–Boveri betatron. The percent depth ionization (PDI) curves are compared with the empirical relation given by Laughlin’s equation, and an extension of Laughlin’s equation is described. The modified absorption coefficient (MAC) method is introduced to correct the isodose distributions for the presence of inhomogeneities, such as lung, with the actual density of lung and the location of the inhomogeneity taken into consideration. Experimental data from measurements made in a water‐cork phantom are presented, and the results are compared with the various calculated methods.
3(1976); http://dx.doi.org/10.1118/1.594227View Description Hide Description
Thermally stimulated radiophosphorescence has been studied as a means of high‐resolution dosimetry. Small grains of CaSO4:Mn phosphor, embedded in a thin Teflon tape, constitute the dosimeter. The light emitted after irradiation is measured with a photomultiplier coupled to the eyepiece of a scanning microscope. With CaSO4:Mn, the phosphoresence at room temperature is sufficient for measurement after doses in excess of 3000 rads. The spatial resolution of the technique is about 0.2 mm. The method has been tested by measuring the dose distributions from a radium needle and a β‐emitting eye applicator.
3(1976); http://dx.doi.org/10.1118/1.594228View Description Hide Description
An experiment has been carried out in which an ACTA computerized axial‐transverse tomographic scanner was used in a series of scans on a polystyrene phantom. Cavities in the phantom contained distilled water and various concentrations of acetic acid or ferric nitrate in aqueous solution. It is shown that the ACTA numbers generated for the low‐Z acetic acidsolution are proportional to the electron density of the solution, but that such is not the case for the higher‐Z ferric nitrate, where photoelectric absorption is significant. The correlation of scanner numbers with electron density rather than with mass density is discussed and, as an illustrative example, the electron densities of whole blood and blood cells are calculated and the relative value is compared with the relative mass densities of the materials.
3(1976); http://dx.doi.org/10.1118/1.594229View Description Hide Description
The differences in the published information concerning tissue kerma in air vs deuteron energy for the d+Be reaction are analyzed in light of some recent measurements. The reason for the discrepancy is determined to be a lack of electron suppression on the Be target in some earlier measurements, and the relation ln(tissue kerma) =ln(1.356×10−4)+2.97 lnE is found to fit the measured data over the deuteron energy range 11–50 MeV.
3(1976); http://dx.doi.org/10.1118/1.594269View Description Hide Description
Permanent implants of 125I seeds at Memorial Sloan‐Kettering Cancer Center are performed by the method of ‘‘dimension averaging’’ to determine the total activity to be implanted; that is, the activity (in mCi) is taken to be five times the average dimension (in cm) of the target region. A nomograph has recently been developed to permit a rapid calculation of the seed spacing required for a uniform distribution of the number of seeds specified by dimension averaging. The number of seeds is also given by the nomograph. The spacing calculation performed with the nomograph, since it involves the two‐thirds power of average dimension and the one‐third power of seed strength, is well beyond the reach of mental arithmetic. Elongation and shape corrections are included. The nomograph spacing result is valid for an equipartition of measured volume among seeds but must be increased by an easily determined factor if peripheral seeds are considered to define the surface of the target region.
3(1976); http://dx.doi.org/10.1118/1.594230View Description Hide Description