summary of responses re: dose determination from x-ray unit

["Sandy Perle" <sandyfl@ix.netcom.com>]

I have been requested to provide a summary of responses to my 
question regarding determination of dose from an x-ray unit. One 
expert said it is VERY difficult to accurately determine the dose, 
since there are many variables .. Phase of power, stability of the 
power source, etc. Below are a few of the comments, and methods with 
references.

Thanks to all who responded, via E-Mail as well as faxing me 
information.

=================
I have a copy of a reference from Archer et al. (1994) that shows the
Dose in Air at 1 meter per unit workload in mGy/mA-min for a W
anode/Al filtered radiography tube.  The equation of the curve in
which the dose in air per unit workload is on the y-axis and the kVp
is on the x-axis is given in the reference as:

Dose=1.222-0.05664*kVp+0.001227 *kVp**2-3.13E-6 kVp**3

So for example, for a tube at 100 kVp, 100 mA and 1 second the output
looks like 4.962 mGy/mAmin *100 mA * 1.67E-2 min =  (783 mR) which 
sounds like it is right in the ballpark in my experience. However, as 
I am sure you are aware, x-ray tube outputs can vary significantly 
with filtration differences and kVp waveform, so this would be only a 
general approximation.  For a specific application, it would be best 
to measure the output of the tube in question. 
==================
NOTE: I was interested in output from a 200 kVp unit.. the following 
was a clarification to my personal correspondence:

Unfortunately, RADCOMP data only go up to 150KVp; however, in my paper
a long time ago (AJR:149,1987), on which RADCOMP was based, I provided
some curves which provide you with the exponent "n" in the
relationship [Exposure is proportional to KVp to the power n].  n
pretty much assymptotes above 140 kVp, and has the value 1.7 for
average filtration and about 2 for high filtration.  So to be
conservative you could use RADCOMP or NCRP-102 to give you the
exposure value per mA-s at 150 kVp, and then just multiply that by
[200/150]**2 to get the corresponding value at 200 kVp
. ==============

We use an empirical equation given by Edmonds[1]:

Entrance air kerma = 836*(kVp)^1.74 * mAs * (1/T+0.114)/(SSD)^2

where T is the total filtration in mm Al and SSD is the source to skin
distance.

This fits Birch and Marshall's calculated outputs to within +-5%.

For REAL machines, however, you have to divide these values by 1.32.

Overall accuray in predicting the output of real machines, if you use
the correction factor, appears to be within +-5 to 10% in the 50 to
140 kVp range for tungsten targets. If you have baseline measurements
at a given kVp and mAs, the output on individual machines can be
predicted at other kVp, mAs values quite accurately[2,3]

[1] Edmonds, "Calculation of patient skin dose from diagnostic x-ray
procedures", BJR 1994;57;733-734

[2] Harpen, "A mathematical spreadsheet application for production of
entrance skin dose nomograms",Medical Physics 1996;23;241-242

[3] Zamenhoff et al, "An improved method for estimating the entrance
exposure in diagnostic radiographic examinations", AJR
1987;149;631-637

------------------
Sandy Perle
Technical Director
ICN Dosimetry Division
Costa Mesa, CA 92626
Office: (800) 548-5100 x2306 
Fax:    (714) 668-3149
  
mailto:sandyfl@ix.netcom.com
mailto:sperle@icnpharm.com

ICN Dosimetry Website:
http://www.icnpharm.com/dosimetry/

Personal Homepage:
http://www.geocities.com/CapeCanaveral/1205 
http://www.netcom.com/~sandyfl/home.html

"The object of opening the mind, as of opening 
the mouth, is to close it again on something solid"
              - G. K. Chesterton -