Eye Physics Logo






PalmOS PDA Applications: MammoSite


This program requires the shared library MathLib in order to properly decay the HDR source. Click here to download MammoSite Calculator 1.0.2 (308K), last updated 9/8/03. Download: MammoSite.prc.zip (128K).

The MammoSite Radiation Therapy System (Proxima Therapeutics, Alpharetta, GA), pictured above, is a recently introduced high dose rate (HDR) brachytherapy applicator targeted at accelerated partial breast irradiation (APBI). Its purpose is to irradiate the tissue immediately surrounding the lumpectomy cavity. A balloon tipped catheter is inserted into the resection cavity and inflated to a diameter of 4 to 5 cm. The inflated balloon shapes and compresses the tissue adjacent to the cavity into a nearly spherical shell surrounding the balloon. A typical clinical protocol for the MammoSite prescribes a dose of 340 cGy b.i.d. at a distance of 1 cm from the surface of the balloon. Dose to overlying skin is the main dose-limiting factor for MammoSite treatment (1).

Treatment planning requires CT or orthogonal radiographic imaging to digitize the path of the lumen through which the HDR wire passes, and to measure the dimensions of the balloon and its distance from the skin surface. Treatment is delivered by an Ir-192 tipped wire which is introduced into the balloon for the time required to deliver the prescribed dose. A single dwell position at the center of the balloon is often used (2). Optimization of the 3D dose distribution using multiple dwell positions within the balloon to compensate for source anisotropy and to constrain skin dose has also been described (3).

Single dwell position

Multiple dwell positions

An important component of every clinical HDR brachytherapy program is quality assurance (QA). One of the QA recommendations of the AAPM TG59 report (4) is an independent verification on the results of treatment planning. The primary intention of these checks is to intercept common human errors such as typographic mistakes during data entry. Verification also serves as a check for the less likely instance of problems with the treatment planning system's (TPS) algorithms or calibration data. Verification checks can be done manually or with small computer programs. Either way, it is desirable for the check procedure to be as quick and easy to perform as possible and yet to have a high probability of detecting significant errors.

geometryThe nearly spherical geometry and cylindrical symmetry of the MammoSite treatment lends itself to a simple QA check which employs a single HDR dwell position at the center of the balloon, a single dose calculation point within 2 cm of the balloon surface on the equatorial plane of the balloon, and the AAPM TG43 report (5) formalism. Normally, 340 cGy (per fraction) is prescribed at 1cm from the balloon surface on the equatorial plane. This is also the point at which this QA program calculates the dose.

Although this calculation can be performed in many ways, it is the type of computing task which is ideally suited to the touch sensitive screen and user interface of a personal digital assistant (PDA).

A simplified version of the AAPM TG43 (5) dose calculation formalism was implemented in this program. In the complete formalism, at a point located at (r,θ) in a medium, where r is the radial distance from the source center and θ is the polar angle relative to the longitudinal axis of the source, the dose rate D(r,θ) is expressed as:


where Sk is the air kerma strength of the source, Λ is the dose-rate constant, G(r,θ) is the geometry factor, g(r) is the radial dose function and F(r,θ) is the anisotropy function. The simplified version used in the MammoSite QA program assumes the medium is water, angle θ = 90°, defines g(r) only over the range 1 to 5 cm, and substitutes the simpler anisotropy constant φan (r) where r = 2 cm for the anisotropy function F(r,θ).

To further simplify the user interface, the source calibration is assumed to be noon of the calibration date in local time, and the activity is automatically decayed to the moment at which the QA calculation is performed.

The PalmOS user interface is organized into "forms" which are roughly equivalent to the windows and dialogs we have become accustomed to in the graphic user interface (GUI) of personal computers. Because of the small screen size of a PDA, the UI must be reduced to the bare essentials. For MammoSite QA, the absolutely essential inputs were determined to be the volume and/or dimensions of the balloon, the distance of the QA point from the balloon surface, the prescribed dose and the current activity of the HDR source. The essential output to be compared to the TPS was chosen to be the dwell time of an HDR source located at the center of the balloon, expressed in seconds.

main formThis is the main form of the program. The calculated dwell time is displayed at the top of the form within a virtual bezel. Below the bezel are input fields for the prescribed dose and the source activity. The current date (9/7/03 in this example) is displayed next to the activity field as a reminder that the calculation is for that date. The calibration date of the HDR source (9/1/03 in this case) is displayed just above the current date. A PDA is controlled by tapping its touch sensitive screen with a stylus. For instance, tapping the triangular bumper buttons adjacent to either of the input fields increases or decreases the displayed value.

Below the dose and activity input fields is a vertically scrollable spreadsheet-like table. Each row of the table corresponds to a balloon volume between 28 and 76 cc in 1 cc increments. The volume is displayed in the first column. The second and third columns list the approximate width and length of the balloon taken from standard data provided by the manufacturer. This provides an additional QA check in that the balloon dimensions for a particular volume can be compared to the TPS dimensions. The rightmost 3 columns of the table display the dose rate (cGy/min/Ci) at 3 user definable distances from the balloon surface. The table can be scrolled to reveal additional rows by tapping the triangular bumpers in the lower right corner of the screen or with a miniature joystick-like navigator located near the base of the PDA.

In this example the prescription is for 340 cGy per fraction at 1.0 cm from the surface of the balloon which was inflated with 35 cc of saline (balloon width 3.99 cm). The current source activity is 9452 mCi. The dose rate at a point 1 cm from the balloon surface (ie r = 3 cm) on the equatorial plane is 8.29 cGy/min per Ci, the total dwell time is 260.3 seconds. You can compare this result to a similar calculation for h = 3 cm based on a modified Paterson-Parker linear source table which yields 260.1 secs or the Greenfield-Tichman-Norman linear source table which yields 261.1 secs. You can scroll the spreadsheet to reveal more balloon volumes by using the scrolling buttons at the lower right of the screen or the joystick navigator on your handheld.

The actions required to complete a QA verification are thus:

  1. iconTap the icon of the MammoSite QA program in the PalmOS applications launcher. The application launcher is roughly the equivalent of the Macintosh or Windows PC desktop metaphor. The main form illustrated above will be displayed.
  2. If the defaults are not correct, adjust the prescribed dose (cGy) and enter the source activity (mCi) at the time of implant. The current activity of your HDR source is entered by default each time you enter this form so normally you do not have to change anything.
  3. Scroll through the spreadsheet to reveal the row which is closest to the volume or dimensions of the MammoSite balloon, then ...
  4. Simply tap the cell in that row which corresponds to the prescription distance. The selected cell is highlighted and the dwell time displayed.
  5. Compare the displayed dwell time to the total dwell time calculated by the TPS. In the event of a significant difference, delay the patient treatment until the discrepency can be accounted for.

Fig 3This is the Physics form. Calibration data (AAPM TG43) for your HDR source is entered in this form. If you set the calibration date for your source and have MathLib installed, the program will automatically decay the source to the current date and time and insert it in the "Activity" field on the main form. The source strength field is updated every time you enter or re-enter the spreadsheet page. The calibration time is assumed to be noon, local time. In this example, the current time is the afternoon of September 1, 2003.

All physics data is automatically saved to the PDA's database when exiting the program and automatically restored when the program is restarted. Program maintenance, therefore, only requires keeping the source calibration data up-to-date. Ir-192 HDR sources (halflife 73.8 days) typically start with activities of about 10 Ci and are exchanged for new sources when they have decayed to about 3 to 4 Ci. Following each source exchange you must enter the new calibration date and activity. When the source activity drops below the "low" threshold, an alert is displayed once per session reminding you that a source exchange is due soon and to update the calibration data stored in the PDA.

Fig 4On the Preferences page, you can choose three different depths for which to calculate the current dose rate (cGy per minute per Ci) and the required dwell time to deliver the prescribed dose. TG43 parameters from the literature (6-10) are incorporated into the program for the new Varisource, new microSelectron, GammaMed and Amersham Buchler HDR sources. Parameters for the selected HDR source will become the program defaults.

Fig 5Additional information about the MammoSite can be accessed from the options menu.


  1. Keisch M, Vicini F, Kuske RR, et al. Initial clinical experience with the mammosite breast brachytherapy applicator in women with early-stage breast cancer treated with breast-conserving therapy. Int J Radiat Oncol Biol Phys 2003; 55(2): 289-293.
  2. Edmundson GK, Vicini FA, Chen PY, et al. Dosimetric characteristics of the Mammosite RTS, a new breast brachytherapy applicator. Int J Radiat Oncol Biol Phys 2002; 52(4): 1132-1139.
  3. Astrahan M, Jozsef G and Streeter O, Optimization of Mammosite therapy. Int J Radiat Oncol Biol Phys 2003 (in press).
  4. Kubo HD, Glasgow GP, Pethel TD, et al. High dose-rate brachytherapy treatment delivery: Report of the AAPM Radiation Therapy Committee Task Group No. 59, Medical Physics, 1998; 25: 375-403.
  5. Nath R, Anderson LL, Luxton G, et al. Dosimetry of interstitial brachytherapy sources: Recommendations of the AAPM Radiation Therapy Committee Task Group No. 43, Medical Physics, 1995; 24: 209-234.
  6. Papagiannis P, Angelopoulos A, Pantelis E, et al. Dosimetry comparison of Ir-192 sources. Medical Physics 2002; 29: 2239-2246.
  7. Daskalov GM, Loffler E, and Williamson JF, Monte Carlo-aided dosimetry of a new high dose-rate brachytherapy source. Medical Physics 1998; 25(11): 2200-2208.
  8. Karaiskos P, Angelopoulos A, Baras P, et al. A Monte Carlo investigation of the dosimetric characteristics of the VariSource 192Ir high dose rate brachytherapy source, Medical Physics 1999;26(8)1498-1502.
  9. Angelopoulos A, Baras P, Sakelliou L, et al. Monte Carlo dosimetry of a new 192Ir high dose rate brachytherapy source, Medical Physics 2000;27(10)2521-2527.
  10. Ballester F, Perez-Calatayud J, Puchades V, et al. Monte Carlo dosimetry of the Buchler high dose rate 192Ir source, Phys. Med. Biol. 2001; 46: N79-N90.

Return to: Palm Programs