Imablog Perspectives of a Canadian in the Old/Deep/New/Geographic South: This is where I ramble on about nothing in particular and post a few nice pictures.

job decisions

About 6 weeks ago, I accepted a research associate position at Duke University. A big jump for me, although mostly in a lateral direction.

My current employer, the Radiology Department at MUSC, likes me so much that they're trying to tempt me to stay with a substantial increase in salary.

Should I stay, even though I've already accepted the new job, or go on to newer fresher pastures and new challenges. It's a tough decision to make, more so because I don't think either decision will be a bad one.

On the one hand, my current location is about as ideal as it gets, aside from the constant worry about hurricanes. And the higher salary is very very tempting.

On the other hand, it's Duke University. In my field, it doesn't get much more prestigious than that. Plus I'll be working with old colleagues again, it's a much larger and diverse population, research is something I've wanted to get into more and I'll finally be in an environment where I can start work on a PhD. And did I mention it's Duke?

Oh, decisions, decisions....

Long nights

My wife is spending the summer working in a medical research lab, part of a summer research program at MUSC for undergraduate students. She's got mice to take care of as part of her research project, which means working late during the day, then making another trip in sometime in the evening to do more things with the mice.

So for the last two weeks, and for a few more weeks to come, our daily routine has been:

• Go to work around 8, 8:30
  
• Leave work around 6 (5:30 if she finished early)
  
• Go back in around 9
  
• Back home around 10:30 or 11
  
• Repeat.
  

Long days for her, and for me as well, since I usually go in with her because it's late at night. It's been making for some pretty long work days. Can't wait for it to be finished with soon.

DHEC survey

State inspectors are going to be cruising through the department the rest of the week on one of their bi-annual surveys. This means there are a lot of people scurrying about (myself included) making sure that everything is where it needs to be, and stuff that needs to be posted is posted. And then later on this year, I think there's supposed to be a JCAHO survey going on. Luckily for me, I think I'll be gone by the time that happens.

New toys

The Exploranium GR-135 we've been waiting for finally arrived. It's a radiation survey meter with brains. The thing I like about it most is that if you find something, you can acquire an energy spectrum and the unit will even tell you what it is, if it's something that's stored in its library. It will even send data to the PC via serial cable too. Pretty slick device. Made by Canadians too!

The guys over at Radiation Safety got one about a year ago, and we couldn't let them trump us with cooler toys, so we got one too. It's going to be a fun tool to play with.

Productive day

After being busy with other things the past couple of months, time for me to start whittling down the list of x-ray equipment I need to get tested by the end of the year. Spend a couple of months doing pretty much nothing except testing, then take the next couple of months to work on other projects. I do this maybe 4 or 5 times throughout the year to break up the monotony of doing all the annual surveys. With all the side projects I keep coming up for me to do, I find this works out better for me instead of spacing out all the testing evenly through the year. The breaks between equipment testing give me some time to work on projects and research. Makes getting back to testing easier too, after taking a break.

Joint Commission Jitters

In a few short weeks, JCAHO surveyors will be descending on our institution to check on how well we do things. This will be the second JCAHO accreditation survey I've been through here. Prior to the first one, there was much scurrying and commotion as everyone tried to get ready for the survey. This time around, most of the scurrying and commotion was spread out over the past year and a half. Still, with a few weeks to go, there's more and more activity to get everyone ready.

JCAHO accreditation is a big deal for hospitals. Being accredited and getting a good score on surveys is a good marketing tool, and shows the institution follows a tight set of standards and practices. Plus it's also required to qualifiy for Medicare/Medicaid funding.

Our survey is scheduled for Nov 17. As far as my little portion of the survey, I don't expect to see much activity. My records are in pretty good shape (much better than when I first started here), and I can pretty much account for everything that I'm responsible for. Since I'm not directly involved with anything patient care related, I will be surprised if surveyors stop by to ask me for anything at all.

Bring it on, JCAHO...

Solving problems

Ever have one of those problems where the more you dig into it to find a solution, the more complicated it becomes?

A couple of days ago, I started digging into a problem we were having with one of our computed radiography (CR) readers. Techs were complaining that they were starting to have to use much higher x-ray techniques than normal to get properly exposed images. So, I head over, and run some quick checks with some of my test objects, and narrow it down to the CR reader. Sure enough, the reader in question was producing a lower exposure index (a number that's related to how much radiation the CR plate was exposed to) than an older CR reader next to it.

The next day, I head over there with some more test objects to get some more quantitative data, and confirmed just how much lower the CR reader was responding.

Today, I decided I had better check our other CR readers. They were all relatively new (installed April 2003), and I had data for some of them when they were first installed. The testing is something I usually try to do on a monthly basis, but the summer was a pretty busy one, so I hadn't been able to get to doing my regular tests on them. Much to my surprise, I found all of the other CR readers producing lower exposure index values too. And to make it worse, they were all lower by the same amount. Each of the CR readers runs a set of diagnostics periodically on various systems, which is very useful. A quick check of those results didn't reveal anything significantly wrong that I thought would cause the problem though. So now, instead of trying to figure out what the problem with one unit is, I have to figure out what could be causing the same problem on 5 different units. Could it be the laser assembly? The light guide? A photomultiplier tube?

At first my guess was the laser starting to fail. But how does it happen to 5 different units by the same amount? A bad batch of lasers perhaps? Who knows. The more I dig into it, the more perplexing the problem becomes.

Laying Low

Joint Commission is cruising through the hospital this week. Time for me to lay low and keep a low profile. Not that much of what I do will draw their attention, but the less of it I have to deal with, the better.

Attention: JCAHO has left the building

Joint Commission survey is over, and I hear we scored pretty high with no Type 1 findings. This is a good thing. So we're good for another 3 years. Excellent.

Oops, much work to do

Ooops, with the holidays and last minute rush to get stuff done in December, the AAPM CT Noise Metric TG task I volunteered for completely slipped my mind. Much research and writing to do now before the end of the month.

My AAPM homework

My assignment for the AAPM task group I'm on is to write a background blurb on the effect of reconstruction filters on CT images.

It's always hard for me to get started when it comes to writing stuff like this, or writing papers. Where to begin, what to talk about, how to organize it... Once I can get started, I'm usually OK though. Getting started is the hard part.

So far, my plan is to have a brief overview of the image reconstruction algorithm, emphasizing the filter. Then a discussion on different classes of filters along with how they affect the image.

Nifty new equipment

Today I get to mess around with our newest cath lab acquisition. Actually I did some initial messing with it doing the acceptance testing earlier this month, but now I get to play and have some geeky medical physicist fun with it.

It's a fairly unique (for the US) system and very cutting edge. As far as I've been told, it's the first system to be installed in the US. A Siemens Axiom Artis with a couple of really large magnets from Stereotaxis used for steering a sepcially designed catheter guide wire through the arteries of the heart.

The x-ray system itself is the first flat panel fluoroscopy unit I've laid my hands on for testing. These things are impressively small and compact. Image quality and performance was ok, but not stunning. Noticed some pixellation getting to the smaller mag modes, but nothing severe.

There are a couple of weeks to go before applications training starts and they put it to clinical use, so this is my last good chance play on it without having to stay after hours. There are a couple of things I plan on checking out.

I'm told the magnets (there are two of them) are 0.7 Tesla permanent magnets, but drop down fairly quickly to about 0.08 Tesla about a meter away. Magnetic fields cause charged particles to move in a circular path, and ion chambers work by measuring ionization in air, so the first thing I was curious about was what kind of an effect the magnets would have on my exposure meters. Further complicating things is that the magnetic field produced by these jammies far from static (which causes much havoc with the angiography unit next door). The magnets articulate and move, which is how the catheter guide is steered. I don't really expect a significant effect, but I don't think it's ever been documented. Maybe I'll even get a short technical note out of it to publish somewhere.

Of course any effect on an ionization chamber while the magnets are articulating is pretty moot, because I don't ever expect to be making exposure measurements while the magnets are being used. But like most physicists, I do things to satisfy my curiosity, and this is definitely one of those things.

The last gamma camera

At long last, almost 3 years after the initial purchase, our final gamma camera has been delivered and installed. Joining our department is a new Philips/Adac Argus Epic single head gamma camera which will do a good portion of our routine and planar imaging, particularly for patients coming down in hospital beds.

Tomorrow I should be able to get in and do the acceptance testing on the unit. Looking forward to it. It's been a while since I've done anything on an Argus. Should be pretty easy this time. Only two sets of collimators and one head to do.

Maybe I should start boxing things up

When your chairman and director pop their heads into your office and ask if you're happy in here, it's usually not a good sign.

Around here, space is at a premium, and someone's always after the space you're in. Things are always getting shifted and shuffled around. Our department has a space committee looking at our space and other peoples' space and seeing what we can re-arrange or annex. Up until now, I didn't think my space was in anybody's plans, but now I'm a little worried.

If they give me some bigger space, I think I wouldn't mind. But I don't think I could fit into a much smaller space.

Cutting edge imaging

Just finished doing the acceptance testing of our new 64 slice CT scanner. This thing will generate a whole bucketload of images without even thinking about it. After two days of playing around, testing and scanning various phantoms, I ended up with a little over 3700 images to send to our PACS. Ended up with some very impressive looking reconstructed images showing a lot more detail in the reformatted coronal and sagittal planes than I expected. It's a cool machine, but radiologists are going to need new visualization tools to handle all the images these things can spit out.

On another note, the article about blogging I was interviewed for a couple of weeks ago was published today in The State newspaper. It's a short little fluffy piece with quotes and stuff from a few other webloggers (including me).

Going filmless, the final stage

This week we're making the final long awaited steps toward going completely filmless in our department. Installation of a CD/DVD burning system to put patient images onto CD/DVD instead of handing patients a stack of films is expected to save the department a ton of money in film supplies and printing costs. It's something we've been able to do before on a limited basis, but it's always been very labour intensive. The new system is pretty much completely automated and so far is being very well received.

The other step is the conversion to digital in one of the last bastions of film: mammography. Last week we had the first of 4 digital mammography units installed, replacing one of our conventional film/screen units. This week applications training started. From what I've heard from the residents, images are great and far superior to regular film/screen. A few snafus with printing and workstation workflow, but probably nothing that won't be solved with a little bit of training and tweaking. By the end of the year, our mammography department will be completely digital and we'll finally be almost completely filmless after almost a decade of digital imaging.

They're my toys

At work I have a number of phantoms that I use for various purposes. The techs always know i'm up to something when I roll up with one them. I call them my patients.

There is one set in particular that the techs always give me grief over, the solid and fillable breast attachments that go along with my cardiac phantom. They always get a good laugh when I pull them out to work with. "Playin' with your boobs again?" and other cracks like that.

All in good fun though I suppose. They are kind of unusual tools to work with.

Navigating Grantland Part 1

A few weeks ago I received word that a small research project proposal that I submitted had gotten funding approval from the sponsoring company. Terrific! My first research grant! This is a big event for me. Most of the grant funds will be used to purchase another phantom, but first there's paperwork to be filled out. And I'll have to navigate the hospital's research and grant administration process. Fine, no problem. There are plenty of people around here with lots of experience in this kind of thing.

First snag I encounter is that I learn is that the hospital takes 25% off the top for overhead. Ok, nobody told me about that when I was coming up with a budget for my proposal so I might have to go back and say I'll need more money. But the people helping me with this tell me that I can probably get an exemption from this overhead charge. Good.

Then, I discover that because I don't have a faculty appointment I can't be the principle investigator (PI) for the grant. I guess the hospital doesn't want to take research money brought in by the unwashed and unlearned masses. But there's yet another exemption that can be applied for that will take a couple of weeks for approval. Fortunately the physician I'm going to be working with is faculty, so he can be the PI instead. I'll just be the one doing all the work. Maybe I'll be able to call myself a co-PI. In any case, I don't think it'll be a big deal. Sounds like it's just one of those administrative things. My project, my paper/poster/abstract. First author is what it's all about.

So now all I need to do is finish up responding to the questions from the sponsoring company, wait for the paperwork to be filled out and hopefully I'll be able to start this project soon.

It never fails...

As soon as you send your exposure meters out for calibration, there are at least 2 pieces of new x-ray equipment that need acceptance testing the next week.

Multi-detector CT and PACS storage

For the past few months, I've been digging through our PACS database extracting all sorts of gory stats on the storage requirements for CT images. After the first of our 64 slice CT scanners was installed and hearing early reports about 1000 slice studies being stored, I started to get curious about the impact of multi-slice/multi-detector CT (MDCT) scans on our PACS archive.

Our PACS archive goes all the way back to 1996 when it was first installed, so there were a lot of numbers to go through.

Here are some graphs summarizing some of what I've found so far.

This is the number of studies per month being stored to the archive. Pretty boring, mostly linear growth happening here.

This is the total number of CT images being stored to the archive each month. You can see fairly steady growth early on with a hint of increase starting around 2000, when the first of our 2 and 4 slice CT scanners were getting installed. The big jump in images/study/month starting around 2003 marks the start of five 16 slice CT scanners being installed. You see another slight increase in the number of images/study/month happening around the end of 2004, which marks the installation of the first of two 64 slice scanners, replacing two of the 16 slice scanners.

Above is the average number of images per study per month being stored to the archive. You see similar trends going on here as with the total images. With MDCT, you have the ability to scan much larger volumes in the same amount of scan time, so naturally you can generate more images. And in order to generate nice looking coronal and sagittal reformats, you need to scan with thin slices which means even more slices.

This shows the maximum number of images being stored each month. As you can see from this graph, we are well past the point where thousand image studies were routine and are getting to where some of the largest studies are approaching 10 000 images. Of course the majority of these images are actually reformats and reconstructed images, not actual 'x-ray on' scans. The jump in max images around July 2003 coincides with the installation of 4 16 slice scanners.

Looking at just one CT room, we see what happens here when the switch is made from 16 to 64 slices. The initial data point is an artifact caused by the vast number of images I sent to the archive from acceptance testing the 16 slice scanner. The unit was replaced with a 64 slice scanner Sept 2004, where you see a slight increase in the average number of images stored each month. Then a few months later once the 'Wow' factor has kicked in and people start seeing just what the machine is capable of, zoom goes the average images per study.

A similar thing, although much more dramatic happens when we look at the total images stored each month for the same room. Following the switch to 64 slices, the total number of images doubles from what was being stored with the 16 slice scanner, partly because of an increased number of patients being scanned, and partly because of more images per study being acquired.

So what does all this mean? With each CT image taking up at least 512 kB a single CT study starts reaching the GB range for storage requirements. This makes the size of the on-line PACS archive in terms of the number of studies it can store becomes smaller. Studies get moved off-line onto tape or MOD sooner. The more MDCT scanners an institution has, the more pronounced this becomes.

Radiologists need new tools for viewing these massive data sets. 3D and volumetric visualization, which used to be considered just eye-candy for referring physicians, becomes essential for making it through the sheer volume of data.

Interesting stuff. I need to write this up and publish it some place.

New diagnostic x-ray system performance standards

Just learned today that the FDA issued the latest revisions to 21 CFR Part 1020, which outlines FDA standards for how diagnostic x-ray machines should perform.

Haven't had a chance to go through it yet, but glancing through the summary, I see a few notable changes:

• Switch to using KERMA and air KERMA for exposure measurements
• Changes to minimum HVL requirements
• Last image hold requirement
• Dose information display to the operator

It's a long 46 page document that will probably take a while to go through, but the summary provides references to the sections that changed so there's probably no need to wade through all of it. Some of the comments might be interesting though.

That's some fertilizer they use

In less than a week, two floors of the corner of one building for the new hospital sprouted up with pillars to support the third floor going up as I left work today. At this rate this first building should be up within a year or so. From what I've heard, this building is supposed to house the diagnostic and treatment building: lab, radiology, offices, clinics etc. Should be fun watching the tower building go up. The campus is going to be a busy place for a long time.

It's always nice to see a project come together

A few years ago the department director had me work on a project to replace our paper radiology request forms. The problem with the forms were numerous: sloppy handwriting, eaten by the fax machine, faxed to the wrong number, duplicate requests, etc. I had set up a couple of web-based databases already, and word got around that I knew how to do this kind of thing so I got 'volunteered'.

So I created a prototype system that was essentially an on-line replacement for the paper forms. PHP/MySQL powered with everything stored in a database (for future datamining/auditing) and requests being emailed out for legibility. Then it languished for a while because we/I didn't have enough resources to develop it further and lacked the political power to get help developing it or to get people to use it.

Then a resident committee, tasked with finding ways to make residents more efficient in light of the new limitations on resident work hours, said "We spend way too much time filling out and tracking radiology requests. We need a better system". So a new committee was formed to come up with a solution. Out came my prototype for a few demonstrations and discussion. With the backing of hospital administration and medical staff, we were able to take my prototype to IT and get someone with more skills than me to create a production version.

Now a year later the project has gone from my initial PHP/MySQL based prototype to a JSP/Tomcat based solution developed by the IT guys (because they have the resources, skills and know how to connect to other hospital databases that make it more useful) and is being used throughout the hospital now. Even though I've had fairly minimal involvement with developing the current incarnation, I still get a sense of pride at seeing how well the product has been received.

It's like seeing a kid grow up and become successful :)

New toy to play with

Just received a much anticipated phantom today: Phantom Lab's Liqui-phil organ phantom. I gotta say, it's pretty schweet too. Purchased with my first grant, I was expecting to get it a month ago. It's here now, and I'm really looking forward to using it.

At first glance it looks rather daunting. You're confronted with a phantom with lots of small and large bits. There are 6 fillable organ chambers (liver, 2 kidneys, stomach, spleen, pancreas) each of which can be positioned anywhere you want in the main chamber. There are also a number of fillable spherical and ovoid chambers that can be inserted into the main chamber, or into the organ chambers to simulate hot or cold spots.

This is going to be so much fun to play with.

Next step: Learn how to fill and insert all the chambers without making a radioactive mess out of myself.

More toys

The latest addition to my collection of phantoms, the Hoffman 3D Brain Phantom from Data Spectrum showed up at my desk yesterday. It's got some refinements compared to an earlier version I worked with several years ago that make it easier to reassemble if it gets taken apart.

It's a pretty cool phantom made up of a bunch of different plates with patterns cut out to simulate radionuclide uptake in the brain. Looking forward to doing a few projects with this one (that list just keeps getting longer and longer).

Paper writing time

It's not often I get a chance to write a paper for submission to a journal or meeting. But when I do, I always seem to be torn between choosing to write it using plain old MS Word, or going with TeX/LaTeX. I've always liked look of TeX generated documents. Problem is I've learned enough TeX/LaTeX to be able to generate some simple equations, but still haven't learned enough to make a full fledged document. A couple of TeX books sit on my shelf to help me learn more (one of them seems to be missing...), but it's always a slow process, because some of the things I end up wanting to do always seem to be things that have a hard time finding solutions to.

And then it comes to crunch time, and usually I end up falling back to MS Word, because it's quick and easy. Just doesn't look quite as nice or sophisticated though.

This time, though, I'm determined to do it in TeX. I have a working TeX environment in the form of MikTeX and what appears to be a decent TeX editor in TeXnicCenter (used Emacs before).

Time to go get started...

MUSC Pumpkin Carving Contest 2005

This year's pumpkins were on display today. Not as many as last year, but there were a few pretty good ones. This year's 1st place winner was the Blood Bank. Surgery's pumpkin was kinda cute.

An early Christmas at work

It's been a good year at work. I've managed to get the department to purchase almost all the things I put on my wishlist this year, and now my office is almost overflowing with toys and equipment waiting for me to find time to play with.

The latest delivery was a much anticipated RTI Barracuda survey meter. This is something I've had on my wishlist for a long time. It comes about as close to a 'do it all' device for us diagnostic physicists as anything else out there. The real reason that I really wanted one this year is because of this: the CT Slice Detector. It's a 30 cm long chamber that's supposed to provide dose information at each point along the length of the chamber. Instant dose profile information. Whoops, spoke too hastily without really knowing how the chamber worked. So it's not exactly what I thought it was, but still gives the same information. After a little more reading about the CT-SD16, I discovered it's actually 1 long chamber with two separate solid state detectors separated by 16 cm. Dose profile information is obtained by doing a regular helical scan with the chamber in the phantom so that the detectors are advanced through the beam. Seeing as how I have the annual surveys for our CT scanners scheduled for the next few Friday's, the Barracuda came just in time too. I'm so looking forward to playing with this new CT chamber. Especially can't wait to try it out on our flat panel CT scanner. Dose profile on the flat panel CT isn't going to be as easy as I hoped, but I'll figure out something to make it work.

Image intensifier distortions

One of the things I've been working on lately is collect images to fill out my image artifact gallery. Today while acceptance testing a new C-arm, I had a chance to demonstrate what happens when large hunks of metal get moved past or near the image intensifier.

All conventional vacuum bottle based image intensifiers are prone to distortions caused by magnetic fields or changes in the ambient magnetic field (which is what large metal objects do), changing the path of the electrons as they travel from the input photocathode to the output window. Normally, image intensifiers are shielded from any stray magnetic fields by mu-metal. It usually works pretty well, but mu-metal can only do so much. IIs will still be prone to distortion when large enough hunks of metal get moved by.

Below is an image of a linear grid phantom (click on the images to see a larger version in the teaching file gallery). All the lines should be pretty straight and perpendicular.

Below is the same phantom, but now a small stretcher has been placed next to the image intensifier. If you look carefully, you will see that now the lines look slightly S-shaped.

Here a much larger stretcher with more metal in it has been placed next to the II. Now the lines are much more distorted and more S-shaped.

If the large metal object is moved past the II during fluoroscopy, you see this cool twisting distortion effect going on as the object gets closer, and then returning back to normal as the object moves away. If you've ever been working on a CRT monitor when a large metal object moves by fairly closely, you will notice a similar phenomenon.

Nuc Med tech wanted

Any nuclear medicine techs out there looking for a job in a well staffed reasonably state-of-the-art teaching hospital? We have an opening for a PET/CT technologist in our Nuclear Medicine department. Dual RT and NM certification is required.

Apply at https://www.applymuscjobs.com/. Click on Search postings and select Radiology Services in the Division dropdown to find the posting.

My first conference presentation

Good: Today I got asked by one of our rad techs to speak at the SCSRT meeting coming up in about 6 weeks.
Inconvenient: I need to come up with a topic and outline by tomorrow so the organizers can send all the info in to get the meeting approved for CE credits.
Good: I can speak on whatever topic I want.
Bad: I don't know what I want to talk about. Current ideas are digital detector technology and identifying image artifacts.
Bad: Whatever I end up talking about will mean lots of work preparing and researching.
Good: I can get Cat 1 CE credits for doing this.
Bad: It goes on the list with all the other things that need to get done in a very short time period.
Good: Something else to add to my CV.
Good: I can probably use this as a MOC SDEP

With my luck I'll be the last speaker on the last day of the meeting.

Four words that are never good to hear

"We've had a misadministration"

As a diagnostic medical physicist, you don't hear those words very often (unless you're talking to a therapy physicist), nor do you want to hear them very often. Not because something really really bad has happened, but mostly because it involves a lot of paperwork and calculating and almost always happen when you're in the middle of doing other things that need to be done yesterday.

Misadministrations in diagnostic radiology don't happen very often because, well, it's really hard for an event to qualify as one. Most of them tend to happen in Nuclear Medicine. 10 CFR Part 35 Subpart 3045 lists a bunch of criteria that have to be met in order for an event to be considered reportable (as far as the NRC is concerned).

10 CFR 35.3045 says (in part):

A licensee shall report any event, except for an event that results from patient intervention, in which the administration of byproduct material or radiation from byproduct material results in--

• A dose that differs from the prescribed dose or dose that would have resulted from the prescribed dosage by more than 0.05 Sv (5 rem) effective dose equivalent, 0.5 Sv (50 rem) to an organ or tissue, or 0.5 Sv (50 rem) shallow dose equivalent to the skin; and
• The total dose delivered differs from the prescribed dose by 20 percent or more;
• The total dosage delivered differs from the prescribed dosage by 20 percent or more or falls outside the prescribed dosage range; or
• The fractionated dose delivered differs from the prescribed dose, for a single fraction, by 50 percent or more.
• A dose that exceeds 0.05 Sv (5 rem) effective dose equivalent, 0.5 Sv (50 rem) to an organ or tissue, or 0.5 Sv (50 rem) shallow dose equivalent to the skin from any of the following--
• An administration of a wrong radioactive drug containing byproduct material;
• An administration of a radioactive drug containing byproduct material by the wrong route of administration;
• An administration of a dose or dosage to the wrong individual or human research subject;
• An administration of a dose or dosage delivered by the wrong mode of treatment; or
• A leaking sealed source.
• A dose to the skin or an organ or tissue other than the treatment site that exceeds by 0.5 Sv (50 rem) to an organ or tissue and 50 percent or more of the dose expected from the administration defined in the written directive (excluding, for permanent implants, seeds that were implanted in the correct site but migrated outside the treatment site).

The radiation dose requirements (> 0.05 Sv (5 rem) effective dose equivalent, 0.5 Sv (50 rem) to an organ or tissue, or 0.5 Sv (50 rem) shallow dose equivalent to the skin) are almost always never met because diagnostic doses simply never get that high. The only procedure in Nuclear Medicine that comes close would be a therapeutic I-131 administration. There aren't any diagnostic radiology procedures that should come even close to these exposure levels. So unless that radiation dose threshold is met, even if the other criteria were met (wrong drug, wrong site, wrong patient, wrong dosage) it still wouldn't be considered a reportable event. It might still be considered a misadministration as far as internal hospital/departmental policies are concerned, but nothing that the NRC needs to be informed about. Still, that doesn't mean records don't need to be kept on the incident. And there are lots of things that need to be done: dose estimate calculations, physician notification, patient counselling, the paper trail, and plans to prevent recurrence. Usually it means getting lots of people involved. Fortunately for the patient though there usually aren't any significant side effects resulting from any misadministrations that happen in radiology. Radiation therapy misadministrations on the other hand, are a totally different matter.

Our new PET/CT is up!

The last week or so has been pretty hectic trying to get everything ready for our new GE Discovery ST PET/CT scanner. Yesterday after a bit of unexpected downtime in the morning, we did the first two clinical patients which went smashingly well. Four more patients on the schedule for today. Lots of work for me to do still on the unit for acceptance testing. Probably should have gotten it done last week, but there were plenty of other things going on with the scanner to keep me busy. Looking foward to finally getting the NEMA PET tests started tomorrow and Friday. In the meantime, I get to go acceptance test my first DR unit tomorrow morning. Should be an interesting experience.

First DR unit is in!

Did the acceptance testing of our first digital radiography unit a couple of days ago, a GE Revolution XR/d.

As imaging technology changes, I usually have to adapt my testing methods to fit. Some technologies render certain tests obsolete or irrelevant while other tests need to be modified, or the analysis changed. In the past, I've had to modify my test and analysis procedures for CR units and more recently multi-detector CT scanners.

This new DR unit was no exception. Being a digital unit, a few things went a lttle faster and easier. Images pop up within 15 seconds of the exposure, so a lot of time gets cut out waiting for the images to appear. The table detector is electronically coupled to the location of the tube and slides along as you move the tube along the table, so no need to fuss with centering the tube over the detector. The folks at GE were even kind enough to incorporate a patient entrance dose display and cumulative exposure counter on the workstation. For some reason though, they've apparently decided to forego any kind of exposure index indicator - some kind of indicator to the technologist that the x-ray exposure they just made falls within an acceptable range for image quality. At least there wasn't one that I could find or that the service engineer knew about.

An exposure index is a very useful tool for providing feedback to the technologist. With conventional film/screen, the tech can easily see whether the exposure was too much or too little by how dark the developed film comes out. With digital imaging, there's no relationship between the appearance of the image and exposure adequacy except in the appearance of noise. Almost all CR manufacturers have some form of exposure index that's displayed to the tech. I'm puzzled as to why this GE DR unit doesn't have anything. Maybe I'll just have to dig deeper to find it.

The first problem this caused was just how to test the kV and thickness tracking for the phototimer. The phototimer is responsible for making sure the image receptor (film, CR cassette, DR receptor) gets enough radiation to produce an adequate image. For film, you measure the optical density (OD). With CR, I use whatever exposure index is provided by the CR vendor as an analog for OD. With this GE DR unit, there wasn't anything immediately obvious to use. So after a bit of mucking around with the software to see what I could find, I eventually ended up using a central region of interest to get the mean pixel value from the raw unprocessed image and tracking that value. Everything seemed to come out ok, although I have no feel for what an acceptable range would be. Something I'll have to work out I suppose. In the meantime, this lack of any kind of exposure index seems like a potentially serious issue as far as providing feedback to the technologist.

The other new thing that needs to be done is the detector evaluation. These detectors need to be properly calibrated, and I'll probably have to include procedures for verifying the calibration. Somewhere in the world of AAPM subcomittees and task groups, there was one putting out a report on testing CR and DR units, which is something I've been waiting a while for and is just what I need for this task. I didn't see it on the list of active task groups, and last I heard the final report was coming RSN, so hopefully I'll see something soon.

We've got a couple more DR rooms being installed in the next few months (hopefully), so I'll have a chance to try out some new procedures in a little while.

No SNM for me this year

Darn, the abstract I submitted for this year's SNM meeting didn't get accepted. No trip to San Diego for me this year.

Still working on the accompanying paper, so maybe I can get that published somewhere.

This is the abstract that I submitted:

Verification of I-131 Half Life Obtained by Conjugate View Measurements Using Medium and High Energy Collimators

Objective: Residence time measurements obtained by serial whole body conjugate view imaging are commonly used in patient specific dosimetry for radioimmunotherapy (RIT) applications. In order to determine the effect of collimator selection on residence time measurements for I-131, the accuracy of I-131 half-life measurements using multiple gamma camera and collimator combinations was investigated.

Method: Serial anterior and posterior whole body images were acquired over a period of 15 days using 4 different gamma cameras with medium and/or high energy collimators. Background corrected geometric mean counts from the images were fitted to a mono-exponential curve to determine the half-life of I-131 for the different gamma camera/collimator combinations.

Results: An average half-life of 8.15 days with a standard deviation (SD) of 0.07 days was obtained from all camera/collimator combinations. A half life of 8.12 days (SD 0.11 d) was obtained for the high energy collimators, and 8.18 d (SD 0.04 d) for the medium energy collimators. These values are all very close to the 8.021 day I-131 half-life from the NuDat 2.1 database. Similar results were also obtained when looking at the measured half-life for single head gamma camera configurations (mean half life 8.15 d, SD 0.12 d). The variation in the therapeutic I-131-tositumomab dose resulting from the differences in measured half-ilfe ranged between 69.8-70.4 mCi.

Conclusion: There is no significant difference in I-131 half-life and residence time measurements made using medium or high energy collimators in dual head or single head imaging configurations.

New beast in the lab

There's a new beast sitting in the radiopharmacy lab at work called a BVA-100 that's supposed to revolutionize the way blood volume studies are performed. The old way is a labour intensive process that took about 3-4 hours to complete. This new machine can do it in about 60-90 minutes with 6 blood draws. I haven't had a chance to familiarize myself with the entire procedure, but it looks pretty simple and is a lot less work. Patient is injected with about 20-40 μCi of I-131 HSA and 6 blood samples are drawn at regular time intervals. Blood is spun down, plasma removed and put into 2x1mL test tubes which are placed in the machine. The computer takes care of getting all the samples counted and spitting out the results.

It's pretty neat and is supposed to be helpful for a few different things like finding appropriate treatments for hypertension. The company came to our nuc med radiologists to get us to try the unit out for a 3 month demo period, which considering our current volume of blood volume study requests was hard to justify. Once their sales guys figured out the people to go to were the cardiologists, those guys were all over it and immediately requested us to proceed with the demo. We just got through with the installation and training period and have done 2 or 3 patients so far with good results and positive response from the techs. If it gets a similar response from the cardiologists and spreads to other potential referring docs, the machine might stick around. The one big downside I've been told is that the consumables have to be purchased from the company and is a little on the pricey side. That's something I leave for te administrators to worry about though.

I'll have to see about finding some time to become a little more familiar with this thing.

Checking out new software

Something called Scenium landed in my lap today. No, it's not a new game. It's some kind of image analysis software for doing analysis on PET brain scans. Currently we're doing qualitative analysis on SPECT and PET images using some software from Hermes Medical.

Haven't found a great deal of info on this new Scenium package yet, and what I have found doesn't tell me much about how it works or what it does. I've got a couple of CDs, so maybe there's something interesting on those.

It'll have to wait until I get back from vacation though.

New equipment gripes

I have a few gripes with x-ray equipment manufacturers when it comes to their digital radiography (DR) offerings.

• The technique controls are terrible. If you're going to make the technologists tap buttons on a touch screen to change the technique settings, give them the option of entering values into a text field instead of having to tap tap tap on the screen all the time. I want to go from 60 to 120 kVp without having to tap the screen a million times.
• Give me a way to turn off the stupid collimation mask applied to the images. It's aggravating when I can't see something at the edge of the radiation field because it's blacked out by the mask.
• Give me some kind of reasonably easy to access service mode so I can see the unprocessed images (or at least an image with a linear greyscale applied to it) and enter arbitrary x-ray techniques. Some of the pre-programmed modes are too pre-programmed.
• Include a screen calibration routine for your FSE's and make it part of the installation and PM procedures. It's very annoying when you tap on a button and the computer thinks you've tapped somewhere else.

More is better

Is it really? For imaging equipment it can be sometimes, up to a certain point. For gamma cameras, more heads is better because it shortens the acquisition time. Once you get up to 4 heads though, you start to run into practical limits like weight and head size.

There's a new CT scanner in the house from a big German company that starts with an S. It's not just any ordinary CT scanner though. This one has two x-ray tubes instead of the usual one. I missed the installation phase when the covers were off so I didn't get to peek inside the gantry to see how things were arranged. I would imagine that the tubes are arranged on opposite sides and slightly offset along the Z-axis (going into the gantry) so that instead of covering the regular 20-30 mm beam width, this thing probably covers twice that. That means it can acquire twice the data in a single rotation. It also means potentially twice the radiation dose to the patient.

Apparently a dual tube scanner proved useful enough to turn it into a commercial product. I'll get my hands on it today and put it through a few checks. It will be interesting to see how this thing performs. Once the apps people come through and set up some clinical techniques, it will be very interesting to see how the radiation dose compares to our other scanners. The timing is pretty good, since it's time for the annual survey on the CT scanners anyway. I'll be able to get some current data for the dose comparisons.

The Interviewer

Tomorrow I have the task of interviewing a well known and prominent physicist in my field for a position here.

This is a strange position for me. I've never interviewed anybody before, and usually it would be me in the interviewee spot and him in the interviewer position. Admittedly, I'm a little nervous about what kinds of things to ask him and talk to him about. I have a few things, but I'll need to come up with a more to fill up the 30 minute time slot I have with him.

This should be interesting.