Saturday, February 25, 2006

Who Owns Your PACS?
...An Online Poll

I'm doing a little research and I need your help. Please submit your answer to the poll below, and have your friends, colleagues, enemies, acquaintances, and anyone else connected with PACS do the same. Many thanks for your help! Just click the link below and answer the simple question:

Poll: Who owns PACS at your institution?

Thursday, February 23, 2006

ITL, Again...

ITL yanked their system, and I thought we were done with them. But alas, they have returned, and the "new and improved" version has been installed. Here is the description from one of my partners:

The system is counter-intuitive, has no cross reference tool on spine MR, and wastes screen space. The imaging tools are difficult to use, and multiple functions require using the "center" mouse button, which actually is the wheel. This is fraught with peril. The cursor travels to the imaging screen by moving down (?!?) off of the menu screen. The hanging protocol must be reset for every study. It has slowed me down so far. The touch screen remains, positioned horizontally, and is completely useless and ignored. Its only purpose is to NOT allow you to move closer to the screen without accidentally touching it and opeing a "toolbox" or switching to the next study.

As I told multiple people at (the site) yesterday, Pacs systems have evolved so that they all are fairly similar in interface. The important term is "evolved," meaning that natural selection has emphasized those traits which are most useful and eliminated the ridiculous and counterproductive. Any system that attempts to dramatically alter the interface (as ITL does) between radiologist and study is attempting to reinvent the wheel. And, as again in the case of ITL, square wheels don't roll real good.

Some improvement. ITL just doesn't grasp how the vast majority of radiologists in the world do their job. I would say that this is a case of engineering/IT types forcing what they perceive to be a "better way" upon us, but how they can think this horrendous approach is "better" is beyond me as a former engineer myself. Some of my people have spent hours with their engineers trying to make them follow our workflow. You've heard the old saying about a camel being a horse designed by committee. Well, this piece of junk is a PACS system designed with total disregard to patients AND radiologists. I'd rather have a camel.

The manager of the imaging center begged me to OK the use of this dog because he's bleeding out cash to pay for film. I gave him a lot of grief, asking why he keeps pursuing this dead-end after all my exhortations. He claims that one of the rads who will use it over at the imaging center gave a provisional OK, and the rest "didn't care." I suspect the one fellow who "approved" did so mainly because eliminating the touch screen meant that he didn't have to keep moving his head up and down, a real problem for him since he just had cervical spine surgery. This was taken as tacit approval from the entire group. As an aside, this points out how important personal communication can be, both within and external to your group. One of my guys gave a vaguely positive grunt, and so that is interpreted as being a huge vote of confidence by the entire group. Great.

I have told my people to use it ONLY if it improves their ability to read studies, and does not prove a hinderance. My personal feeling is to make the center choke on this purchase, give the history, but I'll try not to be too vindictive. We shall see how it goes.

Monday, February 20, 2006

Happy 1st Birthday,!!

Neither the cake nor the child are mine, but you get the idea...

To be accurate, the first post went up on 1/29/2005, but I didn't start logging visitors until 2/18/2005, so that is the date I'll use for the "anniversary".

How far we've come in a year. I'm still tussling with Fuji, Agfa, and GE, still trying to get Amicas accepted at more places, and trying to get them to add a few important items, and I'm still at odds with most of the IT departments I work with (or work for as they seem to see it.) I did manage to get rid of ScImage in the past month, and we did get a "guru" on board at last.

Look for more interesting and weird developments in Year 2!

Tuesday, February 14, 2006

PET/CT Comes to Dalai's Hospital!
A Press Release

Dalai's Hospital adds Pet/CT scanner to its Diagnostic Arsenal
Unit is the first one of its kind in the state

Doctor Dalai's Hospital continues its leadership in imaging by adding dedicated PET/CT to its arsenal of diagnostic tools. Dalai's Hospital was the first hospital in the state to introduce positron imaging, initially with coincidence scanning in 1998. Dedicated PET imaging was launched in March 2001 and we now offers the state’s first fixed-site PET/CT. The new scanner combines PET and CT scanner technology to increase diagnostic confidence and improve patient management with faster scan times and higher-quality images.

“At Dalai's Hospital, we are always looking for ways to improve patient outcomes,” said the Hospital's Chief Operating Officer. “We’re glad to continue to lead in this diagnostic area by providing another much needed tool which gives physicians more detailed information to help patients. For us, its part of providing the quality healthcare to our community that Dalai's Hospital is known for.”

Positron Emission Tomography (PET) works by creating images of the biological functions of the body to reveal disease states. Prior to the exam, the patient receives an injection of a tiny amount of radioactive tracer, which emit signals as they travel through the body. Most scans utilize a positron-emitting form of glucose, which allows mapping of metabolism. PET measures the degree of sugar uptake, taking advantage of the fact that cancers use more glucose than normal tissues. The Computed Tomography (CT) aspect of the scanner improves the quality of the PET scan and adds an anatomical basis for localizing where the sugar or glucose uptake has occurred

This new class of PET/CT scanner, Siemens Biograph 16, is the first to utilize the proprietary lutetium oxyortho-silicate (LSO) crystals plus true 3D acquisition to allow faster acquisitions of images while improving image quality. No other manufacturer offers this critical technology. For patients, this means more comfort and confidence in the treatment of their cancer and other disease states.

Physicians believe that the improved images produced by the PET/CT could reduce the number of invasive procedures required during follow-up care, including biopsies and even unnecessary operations. “The quality of the images, and the added anatomic dimension, gives us greater ability to find and monitor disease,” said Dr. Dalai, radiologist and director of nuclear medicine at Dalai's Hospital.

Dr. Dalai added, “In fact, images produced by the new scanners are so precise that in some cases we have found malignancy that probably would not have been detected with any other technique. The addition of CT to the PET scan also helps us differentiate normal from abnormal tissues to a much greater degree than we could before.” Dr. Dalai is confident that the new combined PET/CT scanners will play a key role in continuing to improve patient care.

In oncology, PET/CT provides for early diagnosis, more accurate tumor detection and precise localization, improved biopsy sampling, and better assessment of patient responses to chemotherapy and radiation therapy. PET/CT also is used by cardiologists to detect certain types of heart disease and by neurologists for assessing disorders such as Alzheimer’s disease.

Sunday, February 12, 2006

"I'll Sleep When I'm Dead"

No, I've never heard of this movie either, but I think it did well in its native UK. Still the title is rather catchy, if disturbing, and introduces today's topic: Call vs. Sleep.

I've written a piece about overuse of imaging in the ERs and another one about night call. Nothing much has changed in the meantime. We do now staff the other hospital I mentioned, and there is a call service in place over there that does a very nice job from 11PM until 7AM. We have voted to extend the service's coverage to our entire practice, and this should be in place in a month or two, depending upon how quickly credentialing can be accomplished.

The whole concept of call troubles me, and it always has. Yes, I realize that people get sick in the middle of the night and require immediate care. No question there. However, the American public has been conditioned to come to the ER with any and every twinge, and once there, the ER has to X-ray, scan, or otherwise bless them with modern medical technology before they leave. There is a significant subset of ER patients known as "frequent flyers" whose charts are longer than the Encyclopedia Brittanica and whose Xray jackets from the old days of film have enough silver in them to make a car. (Of course now in the digital era they could get their jackets as their own personal Greatest Hits CD collection...a 15 disk K-Tel special.) I have taken to mentioning this in reports..."Today's CT of the head for altered mental status is compared to the 20 previous head CT's for the same indication, with the most recent prior study being from...." Bet I get in trouble for that eventually, huh? I have to confess with some degree of shame that I actually begin to get angry about reading these studies, and the more fatigued I get, the nastier I feel about it. A colleague tried to set me straight, if you will, by explaining the ER's attitude: "We do such a great job of diagnosis with a CT that they now rely on us completely." I'm supposed to feel better about that? Do you mean to tell me that medicine has become little more than imaging? No more Marcus Welby, M.D.? Pretty soon, we will be little more than auto mechanics (no offense, but you know what I mean.)

Anyway, a spot-check of ER clients reveals a potpourri of reasons for their selection of 3AM:

  1. The headache I had for the last 3 days got a little worse tonight.
  2. I work during the day.
  3. The ER isn't as busy at 3AM.
  4. The nurse at my Nursing Home just noticed this.
  5. I'm going on a trip in the morning.
  6. I'm scared.
  7. I can't afford a doctor of my own, so I came to the ER.

And so on. Frankly, I think this whole situation is an embarassing indictment of American Medicine. We have failed these people during the day, and so they come back to haunt us at night. I wish I had the answer to fix it. You can bet that if a certain female Senator from New York ever gets back to the White House, she will fix it, and us, with socialized medicine.

But back to the question of call and sleep. I chose the movie title as an opener to reflect the apparent views of lots of doctors in general, and radiologists in particular. Sleep? What's that? Only weenies need sleep. My talents are needed NOW. I'll sleep some other time. Basically, we can't or won't realize that we (and our colleagues) have to sleep. But pushing ones' self to the limit and continuing to work after 24 or more hours of wakefullness is a really bad idea. I can cite article after article after article that shows we make more mistakes when sleep-deprived. It's a no brainer (almost literally!) Somehow, in Medicine, we don't take this for granted as we would in any other industry or situation. DM Gaba noted in a paper presented at the Proceedings of Enhanced Patient Safety and Reducing Errors in Health Care in 1998 that, most high-hazard industries the assumption is that fatigue and long, aberrant work hours lead to poor performance, and the burden of proof is in the hands of those who believe that such work practices are safe. In medicine, concerns over discontinuity of care, and difficulties in changing medical culture have pushed the burden of proof into the hands of those who wish to change the status quo. Given that medical personnel, like all human beings, probably function suboptimally when fatigued, efforts to reduce fatigue and sleepiness should be undertaken, and the burden of proof should be in the hands of the advocates of the current system to demonstrate that it is safe.

The sleep-deprivation starts in medical-school, and was long institutionalized by internship and residency. In the past few years, government regulation has limited resident's weekly hours to 80, and their continuous shift time to 24 hours. This presents somewhat of a problem in staffing, and there is indeed an impact upon the continuity of care by the residents. Still, does the benefit of continuity override the detriment of an exhausted resident pushing the wrong drug, or falling asleep whilst operating? These things have indeed happened, you know. (Or do you?)

My interest and maybe ire were rekindled by an presentation from Nabile Safday, M.D., cited in RSNA News. Dr. Safday found lots of interesting stuff, like radiologists have good vision by and large with mean acuity of 20/16, better than normal. But here is the part that hurts (italics are mine):

When asked, “Have you ever fallen asleep while driving after being on-call overnight,” 28.5 percent said, “Yes.” Almost half (47.6 percent) said they had nodded off while reading a patient study during on-call hours. Finally, the trainees self-reported that an average of 17.5 percent of errors they made on-call were due to a lack of sleep.

Bottom line: we are hurting EVERYBODY involved by pushing ourselves to the limit in this manner. We are guaranteed to make more mistakes in reading, and anyone who says they are as accurate after being up all night is fooling you, and probably him/herself. I couldn't find any figures on whether any rads or other physicians have actually died in car crashes attributable to sleep-deprivation, but you can bet it has happened.

So why do we do this to ourselves? About 95% of the answer is MONEY. Covering your hospitals, and especially your ER's, is a zero-sum game. Either you do it yourself, you hire more people to do it so you all have call less often, you hire a nighthawk to do it for you, or you hire a call service to give prelims (or final reads). The only currency we have is money and time off. If my partners and I take call and then work into the next day, as we are doing now, we get more money (ignoring for the moment the fact that a lot of ER cases don't pay anything), and we allow somebody else to be off the next day. Our theory has been that you can't enjoy the day off after call anyway, because you are so tired, so you might as well work a half day, right? Wrong, buffalo-breath, but not everyone agrees with me. If you don't want to pay a call service, and you don't want to hire an internal night-hawk, there is, of course, the option of going to shift work. Some of my partners think this is a great idea. One would work a week of 11PM to 7AM, with the day off afterwards. Just wonderful. Except there is one problem...remember the zero-sum calculation. Somebody has to be working during the day for the call guy to be off! So you see, this is yet another form of robbing Peter to pay Paul. You can't get something for nothing, you know.

Personally, I think the internal 'hawk is the way to go, but those birds are few and far between, and they don't seem too disposed to coming to our little town. There isn't much to do here during the day, so why come here to work at night? That leaves the call service, and this is the path we have decided to follow. There are those who are not happy about the cost of $50 per study, and that for just a preliminary report that we will have to reread in the morning. But tell me, what price do we put on our patients' health? Or our own, for that matter? Hmmmmm?

Wednesday, February 08, 2006

The CR Ballad

Come 'n listen to my story 'bout my new CR.
It cost a pretty penny but the pitchers sure were poor.
But then one day we were lookin' at a chest,
And darn if it weren't lots worse than all the rest...
Missed a in trouble...
Well, the first thing we did was to call the service boys,
And they all said, "ain't a problem with our toys,
Any product with our label is as perfect as can be,
It's gotta be yore doggone PACS that will not let you see..."
DICOM problems, or somethin'
So we called up our PACS service boys, and this is what they said:
"We don't mess with any data or with any DICOM head.
What the CR sends to us is what we send on to your station,
The FDA don't let us alter yore interpretation"
So they got out every service guy from east of Tennessee,
And they worked on it for hours just for you and just for me.
And doggone if they didn't make it better than before,
So I don't have to kick that darn machine right out the door.
Gonna keep it, since it works now.
Here's the moral of the story, folks. On this you can depend:
A lot of finger-pointing gets you nowhere in the end.
If yore CR's lookin' crummy and you can't DICOM-Retrieve,
It could be that your current vendor really ought to leave.
Or at least help you out, like ours did. Y'all come back now, hear?

Monday, February 06, 2006

Sunday, February 05, 2006 joins Hitsphere!

This weblog has just been accepted into the HITSphere! It isn't an easy task:

The HITSphere is a network of premium weblogs that write content about the healthcare, medical, and clinical informatics and information technology (IT) industry. Combined, these sites reach a large readership of influential healthcare technology professionals. Learn more.

Many thanks! Be sure to check out the other (much more) illustrious blogs.

Saturday, February 04, 2006

Adult Image Detection

Images courtesy of

Every so often, I succumb to the urge to click an ad on a Google search. This morning, I decided to "scan my computer for unwanted porn". I guess there must be some way to tell the difference between "wanted" and "unwanted" porn. I ran the scan, and about all it found were articles and references to mammography stored on my computer. Wow. (Those were "wanted", by the way.) I wonder if it will report you if you have nekkid baby pictures of your kids on your computer.

The website itself had this interesting note on their technology, called Adult Image Detection: "An advanced algorithm is used to detect adult images by assessing skin-tone levels and distribution." The image above gives you the idea, although I'm not sure the young lady's epidermis was detected by this algorithm, or just with something like Photoshop's magic wand tool for the example.

The process might lend itself to the radiological world. I know there is a lot of work being done in CAD (Computer Aided Diagnosis) and such. I use one form of this for mammography, but honestly, the machine misses more than it finds, and flags stuff that isn't there. Maybe something simple like looking for contiguous bands of density (like a subdural bleed) would be a start.

Just a thought. You never know where you might find ideas for new approaches to things.

Thursday, February 02, 2006

8000 Visitors!

Visitor 8000 comes here via Cox Communications in San Diego, California. Isn't it amazing that yet another thousand folks came here to look around with very little new content? I'm actually rather ashamed of the slow progress, but that will change. We are in the process of building my own personal PACS (OK, my group is paying for it, but I still like to think of it as mine.) Stay tuned for all the gory details...