Journal Club: A new look at CT dose measurement: Beyond CTDI

Hope whoever was reading was amused by the last journal club article.

Figuring out how to measure CT does is getting more and more difficult, with the increasing popularity of multi-slice helical CT scanners that can now acquire upwards of 64 slices per gantry rotation and with beam widths approaching 20 or 30 mm. Current methods of measuring CT dose (CTDI) date back to when single slice scanners were the only thing available. So perhaps it’s time for something better and more accurate. As long as it’s easy to do and doesn’t require much more in the way of extra equipment.

Dixon RL, “A new look at CT dose measurement: Beyond CTDI”, Med Phys (30), 1272-1280 (2003)

Abstract:

Equations are derived for generating accumulated dose distributions and the dose line integral in a cylindrical dosimetry phantom for a helical CT scan series from the single slice dose profiles using convolution methods. This exposition will better clarify the nature of the dose distribution in helical CT, as well as providing the medical physicist with a better understanding of the physics involved in dose delivery and the measurement process. Also addressed is the concern that as radiation beam widths for multi-slice scanners get wider, the current methodology based on the measurement of the integral of the single slice profile using a 10 cm long ion chamber (CTDI100) may no longer be adequate. It is shown that this measurement would underestimate the equilibrium dose and dose line integral by about 20% in the center of the body phantom, and by about 10% in the center of the head phantom for a 20 mm nominal beam width in a multi-slice scanner. Rather than making the ion chamber even longer to collect the broad scatter tails of the single slice profile, an alternative to the CTDI method is suggested which involves using a small volume ion chamber, and scanning a length of phantom long enough to establish dose equilibrium at the location of the chamber. With a modern CT scanner, such a scan length can be covered in 15 s or less with a helical or axial series, so this method is not significantly more time-consuming than the long chamber method. The method is demonstrated experimentally herein. ©2003 American Association of Physicists in Medicine.

Easy access to 6 years of Medical Physics

During the course of doing literature searches, I frequently find myself looking for back issues of Medical Physics. That means looking them up online or making a trip to the library to dig up the articles I’m looking for.

Fortunately, there are ISO images of Medical Physics availble for 1997-2002 (hopefully 2003) soon, which you can download and burn onto CD (and which I have done already). Then you have to remember where you put those darned CDs. And if you’re anything like me, that ends up being in the desk drawer with a bucket load of other things, and they end up getting all scratched up.

Then I stumbled on a couple of articles about building a virtual CD jukebox. What a fabulous idea!! And it’s so easy to do! So that’s just what I did. Took a while to FTP over a gig worth of ISO images to my Linux server, but once they were there it was a piece of cake.

And thanks to Samba, I have quick and easy access to 6 years worth of Medical Physics mapped to a network drive on my Windows computer. If I need to go any further back, I’ll have to go online or to the library, but usually I’m not pulling anything from much later than 1997 anyway.

I suppose getting the article I’m looking for online isn’t much more difficult. Just a few clicks away on the website. The virtual CD jukebox method doesn’t really get me any more ease of use, but it is neat, and I did learn something from it. This has a higher geek factor I think.

Journal Club: Characterization of Organic Illumination Systems

This paper comes out of the Western Research Lab at Compaq: Characterization of Organic Illumination Systems. Making organic things emit light is a very trendy topic now, with a good deal of research going into OLEDs (organic light emitting diodes) and similar devices. This paper from way back in 1989 is an interesting read on some early investigations on the subject. They do some very interesting investigations on materials you find in most kitchens.

Abstract:

Recent anecdotal reports of novel principles of illumination have stressed qualitative aspects. This note presents a quantitative study of an organic illumintation system, characterizing the temperature and current-flow properties of the system as functions of time and device parameters. Theoretical and practical implications of these measurements are discussed.

Wrestling Samba

Samba is a very cool tool, especially if you have a mix of Windows and *nix servers and clients. But it seems like every time I have to do anything with Samba, it’s like pulling hair and teeth to get everything to work properly. On paper, Samba is easy to set up and configure. In real life, I always end up wanting to pitch the Windows box out the window.
You read through the Samba man page, set up smb.conf to do what you want and then go to the Windows box and try to map the share you just set up. Then all you get are message boxes from Windows claiming the username and password are invalid. Or it works fine on one Windows box, but doesn’t work on any other ones. Or only one account works, but none of the others. Nothing wrong with what you put into smb.conf. Works just fine on the server with smbclient. Just the stupid Windows boxes refuse to work properly.
Argh.
Update: Ok, finally got things to work nicely. Upgrading to Samba 3.0.2 did the trick. I guess the docs I was using and the Samba version I had installed didn’t quite match up. Still want to pitch my Windows box out the window though.

Kickass server admin tool

The other day while setting up Samba on our RH 9 box, I found out about this very cool sysadmin tool called Webmin. This thing is absolutely amazing. An entirely Perl/web-based application for helping with server sys admin. The primary purpose appears to be to provide a nice usable GUI for configuring and setting up the vast array of server software you will find on a typical server. It’s all module based, so there is a module to help you deal with Apache config, setting up Samba and Samba shares, NTP, Sendmail, Procmail and more. All kinds of good stuff. No more wading through config files and trying to remember where the config file for whatever daemon is located.

Of course, you still need to know what you’re doing, and not every single configuration option for every daemon or config file is available. This is no substitute for sys admin knowledge. It just places all the config stuff in one easy to access location and provides an easy to use interface. Very sweet indeed. And best of all, it’s open source!