Sunday, February 13, 2011

Little Black PACS Station

Cyril Kornbluth, one of the most beloved Science Fiction writers of the '50's, died of a heart attack at the rather young age of 34. Perhaps his most famous work was the short story, "The Little Black Bag":
"The Little Black Bag" was first televised as a live act on the television show "Tales of Tomorrow" on May 30, 1952. It was later adapted for television by the BBC in 1969 for its Out of the Unknown series. In 1970, the same story was adapted by Rod Serling for an episode of his Night Gallery series. This dramatization starred Burgess Meredith as the alcoholic Dr. Full, who has lost his license and become a derelict. He finds a bag containing advanced medical technology from the future, which, after an unsuccessful attempt to pawn it, he uses benevolently — reclaiming his career and redeeming his soul...but not that of the guttersnipe he takes in as his receptionist/assistant.
The text is available at Gutenberg Project Canada. Here are the 1952 (Tales of Tomorrow) and the 1970 (Night Gallery) adaptations:

Gotta love Dr. Fullbright's persistent cigarette in the 1952 episode, and the different interpretations of the "guttersnipe" in both clips.

With all due respect and gratitude to Cyril, may I present my own (per)version of the story, The Little Black PACS Station:

Dr. David Fullerbrush was a radiologist, although he was not a particularly good at it. Which probably explained why he was working for ImagingTemps, the least-respected locums agency in the country. His current assignment was equally admirable, covering general radiology at the county hospital in Teaneck, New Jersey.

The job wasn't that bad, all things considered, for a 62-year-old who had finally learned to read CT scans five years before, and couldn't begin to differentiate a bleed on an MRI of the brain from a normal thalamus. This was, in fact, how he found himself working for ImagingTemps. It seems that he declared a growing opacity on chest radiographs a "normal cardiac silhouette" for five years running, a finding which didn't quite agree with the autopsy results of a cantaloupe-sized tumor. IT's policy, fortunately, was "don't ask, don't tell, don't check credentials," and they were usually on time with Dr. Fullerbrush's salary. He did end up working some colorful inner-city gigs, and he never stayed at any one site long enough for him to hurt too many patients.

Teaneck County was the sort of place that you might not want to visit at the wrong time of day. In fact, most of the physicians and staff wouldn't go out the door after twilight, as it was considerably safer to stay inside until dawn. It was the hospital of choice for the local knife-and-gun club, and when one of their members rolled into the ED, it was a certainty that friends of the patient would shortly inspire a member of one of the rival clubs to make a similar visit.

Having graduated from film to PACS only a few years before, Teaneck County was ahead of many of Dr. Fullerbrush's placements. The system, a bargain from Typical Electronics, actually did work, although it tended to choke on the 5,000 slice whole-body/head/total spine/triple extremity CT with head-to-toe CTA that was de regeur for all trauma patients. Fortunately, Dr. Fullerbrush was familiar with this piece of equipment, as TE was rather ubiquitous in the sort of hospitals he frequented.

But one night, about two weeks into his tenure, he entered the dingy reading room and found all the stations filled by the other ImagingTemps employees doing time in Teaneck. They barely acknowledged him, as they were trying desperately to read at least one part of an exam before their station crashed. David looked around, shrugged, and strode past his colleagues and into an alcove hidden in a far corner, hoping to find a place to sit and doze. In the darkened room, there was indeed a worn but serviceable chair, and he plopped down in it with a sigh of relief. But before he could drift off into the safety (more for Fullerbrush's patients than for him) of slumber, he spotted a faint glow on the battered desk in front of him. He sat up a bit straighter in the chair, and the glow came into focus. It was a monitor of sorts, although nothing like David had ever seen before. It was solid black, rather like a brick of black glass, and it had no markings or other identifying features. Save for the glow, which was an iridescent shade of violet.


Startled, David looked around. There was no one in his alcove, certainly no one who could have spoken in the soft feminine voice that seemed to originate from a spot near his own corpus callosum.


"Who, me?"

"Thank you. Voice print registered, Dr. David Fullerbrush session now active."

"Ummm, OK..." Fullerbrush had dealt with Speech Recognition at those hospitals which placed a higher value on saving the cost of transcriptionists than on accurate reports, but this didn't quite fit the template.

"Would you like to interpret the next patient, Dr. Fullerbrush?" requested the ethereal voice.

"OK, sure, whatever..." replied the wary rad.

"Thank you. Next patient is Ralph Rococo, 39 year old male, status post gunshot wound to the buttocks. Presenting image now."

And with that, the purple glow from the little black slab swirled around a bit, and settled into a three-dimensional, rather solid appearing rendition of a human torso. David sat back, utterly devoid of any thought as to what to do with this apparition. He brought his hands to his face, hoping to rub his eyes and make the image disappear completely. As he did so, however, the floating image twirled around, and the surface melted away on the side facing him, revealing the organs and blood vessels beneath. The more he moved, the more he found he could manipulate the ghostly form.

"Would you like to begin dictation?"

"Uh, sure. Ummm... Whole body CT is performed without contrast. 3D rendering is obtained." Dr. Fullerbrush, nonplussed as he felt, was not about to lose the reconstruction fees. "Ummm... no acute findings. Signed."

"Dr. Fullerbrush, perhaps you would like to mention the tract of the bullet that has entered the patient's buttocks and severed his right iliac artery and vein?"

And within the floating image, a bright yellow line appeared, presumably showing the path of the projectile. Pooling collections of blood, in neon blue and red, became visible at the level of the vascular injuries.

"Oh, yeah, I was about to take care of that. Please include it in the report," David replied sheepishly, hoping none of his colleagues could hear the exchange between him and the voice in his head.

"Would you like to treat this wound?"

"WHAT? I'm not a surgeon."

"Would you like to treat this wound?"

"Yeah, sure. Why not? I'll just wave my pinkie and everything will be fine!"

"Standard protocol requires only your verbal order. Treatment commencing. Artery regrown. Vein regrown. Muscular and cutaneous damage repaired. Foreign body teleported. Treatment complete."

Dr. Fullerbrush sat back with a mixture of awe and horror. Perhaps he should have attended RSNA at least once in the past few years. Certainly TE couldn't have come up with something like this.

Just then, the ED doc burst into the reading room, yelling at the top of his lungs.

"Basically, I have not been seeing anything of this sort in my long and illustrious career! The unfortunate patient was laying on the stretcher with blood pouring out of the wound in his tremendous backside, and then stopped it did! Right there! The bloody damn entrance wound closed like never it had been there! Unbelievable!"

David cowered in the back room, uncertain of what to do next. The pause seemed to trigger something in the black glass.

"Dr. Fullerbrush, I must perform some brief preventive maintenance. Please stand by."

And with that, the slab went completely dark for a few seconds, and then lit up once again. The purple glow again appeared, but this time, there were words solidifying in the display:

Welcome to MacOSM, copyright MMCCCXI.

MMCCCXI? thought David. 2311? NO! This is impossible! But it's here! My God, the advance in technology! How many people could I save with this? This could make up for all of my old faux pas!

", show me the next patient, please."

"Yes, Dr. Fullerbrush. Next patient is Lilly Blanc, age 57, chief complaint diffuse abdominal pain."

And again, the purple light solidified into a spectral body, floating in the air in front of the desk. This time, David stopped to study the image more carefully. He waved his hands with more precision, and laid bare the peritoneum of the patient. He paused, and gasped. Wrapped around the pancreas was a 7 cm mass, displayed in pulsing, glowing mauve. The ugly lesion insinuated about the celiac and superior mesenteric arteries, clearly an inoperable cancer. Multiple glowing orbs, ranging from marble-sized to larger than a grapefruit, could be easily seen within the liver.

"Um, Miss, computer... can we fix this?"

"Yes, Dr. Fullerbrush. Please verbally request treatment."

"Yes, absolutely! Treat the patient!"

"Yes Dr. Fullerbrush. Treatment commencing. Sequencing neoplasm DNA. Tuning optimal resonant destructive frequency. Ablating tumor now. Regenerating normal pancreas and liver tissue. Treatment complete. Shall I rejuvenate the patient as well?"

"Why not?"

"Commencing rejuvenation sequence. Atherosclerotic changes reversed. Articular cartilage resurfaced. Emphysematous lung damage repaired. Biological age regressed approximately 13.7 years."

"That's INCREDIBLE, David!"

Fullerbrush turned around with a start, and was saddened to see the head of ImagingTemps, Nosi Hachoo, standing behind him, jaw slack with wonder. Hachoo was known to pay unannounced visits to keep his minions in line, and it seemed he had chosen this rather fortuitous moment to appear.

"Hachoo, this is some incredible technology from the future. I don't know how it got here, but it could revolutionize medical care. Just think of how many people we could save with this!" David cried.

"Yes, of course, you save people with it, hack radiologist! You're working for me, and so it's MINE. Can you imagine what people will PAY for us to take care of them?"

"No, Hachoo! You mustn't think like that! This belongs to all of humanity!"

"Great! All of humanity can pay me to watch it for them!" Hachoo chuckled.

With that, Hachoo grabbed for the black slab. David lunged after him, knocked him to the floor.

"Computer!" Fullerbrush yelled. "Delete Nosi Hachoo!"

"I'm sorry, Dave. I'm afraid I can't do that," whispered the soft, feminine voice. "An attempt has been made to commit a homicide with this station. License has now been voided. Have a nice day."

The sides of the slab separated from each other, and floated to the corners of the room, fading to nothingness as they approached the walls. Pinpoints of bright light emanated from the core of the slab, and exploded outward.

"Oh, my God!" David cried, as he slumped to the floor. "It was full of stars!"


Friday, February 11, 2011

Third Place....

Oh, well...  Third place (missed Second by ONE VOTE!!!) is certainly better than not even being in the race.  Congrats to the Winner, ScienceRoll!!!

Sunday, February 06, 2011

The First DIAGNOSTIC Radiology Application For Mobile Devices

It's because of MIM Software that I have an iPhone.  I convinced Mrs. Dalai that I needed one since MIM's first app for the iPhone 3G would allow remote viewing of PET/CT's.  Of course, I never actually used it for viewing, but I did proudly show off the fused PET/CT images to anyone who would look.

I've been in touch with MIM since those early days, although I'm sadly still not a full-fledged customer.  Still, I had the opportunity to help out with the final testing and FDA certification of their latest iPhone/iPad app.  Yes, I'm a qualified radiologist.  (Disclosure:  I received a $20 light meter to allow me to check the ambient levels in various parts of the house and outside for the test.)  On February 4, the final approval came through:
FDA clears first diagnostic radiology application for mobile devices
Provides wireless access to medical images for iPhone, iPad users
A new mobile radiology application cleared today by the U.S. Food and Drug Administration will allow physicians to view medical images on the iPhone and iPad manufactured by Apple Inc.

The application is the first cleared by the FDA for viewing images and making medical diagnoses based on computed tomography (CT), magnetic resonance imaging (MRI), and nuclear medicine technology, such as positron emission tomography (PET). It is not intended to replace full workstations and is indicated for use only when there is no access to a workstation. . .

Radiology images taken in the hospital or physician’s office are compressed for secure network transfer then sent to the appropriate portable wireless device via software called Mobile MIM. Mobile MIM, manufactured by Cleveland-based MIM Software Inc., allows the physician to measure distance on the image and image intensity values and display measurement lines, annotations and regions of interest.

In its evaluation, the FDA reviewed performance test results on various portable devices. These tests measured luminance, image quality (resolution), and noise in accordance with international standards and guidelines. The FDA also reviewed results from demonstration studies with qualified radiologists under different lighting conditions. All participants agreed that the device was sufficient for diagnostic image interpretation under the recommended lighting conditions.

The display performance of mobile devices can experience significant variations in luminance levels even between mobile devices of the same model. The Mobile MIM application includes sufficient labeling and safety features to mitigate the risk of poor image display due to improper screen luminance or lighting conditions. The device includes an interactive contrast test in which a small part of the screen is a slightly different shade than the rest of the screen. If the physician can identify and tap this portion of the screen, then the lighting conditions are not interfering with the physician’s ability to discern subtle differences in contrast. In addition, a safety guide is included within the application.
The app takes the thick client approach, or as thick as you can get with the iPhone/iPad, and it does function smoothly.  While it is a bit spartan relative to the full desktop program, but it does provide enough tools to truly perform a diagnostic read.  And that's now legal, thanks to the FDA.

I do have to add that while the app is free, the desktop software required for operation isn't, but for what it does, I think it's price is pretty reasonable.

However, there is another option, as MIM reminded me after this post first went up:

Appreciate the mention in your blog.

I feel obligated to mention that desktop software is not required for its operation. You may recall that you accessed MIMcloud for your clinical study.  The cloud is how most people will use it, and its an inexpensive "pay as you go" option.

We are already pushing to complete MIMcloud 2.0 which makes managing groups and individuals for sites way more convenient, as well as making the MIMcloud interface much more friendly (tricky business encrypting and decrypting data on demand in a timely manner...). We plan for that in the next few months.

People wanting to use MIMcloud now can either manually browse to, and upload files (from their computer) right from the MIMcloud website, or get a hold of a MIM 5.1 beta and try it out that way. MIM 5.1 integrates with cloud and makes it simpler.

For MIMcloud 2.0, we'll also have a simple utility which makes your MIMcloud as easy to access as a DICOM node on your local network. That will boil it down to pure-simple for those users who don't ultimately end up using MIM workstations.

Congratulations to MIM for this milestone.  This may well be the first of many, but it is certainly worthy of your attention.

Being away from your workstation is no longer an excuse.

The 2010 Medical Weblog Awards!

Sometimes self-promotion gets you somewhere., a great site in the spirit of Engadget and Gizmodo, but with a medical bent, has an annual Medical Weblog Award.  This year, I am very proud to announce that Dalai's PACS Blog is up for the award in the Best Medical Technologies/Informatics Weblog.  In fact, I'm first on the list, which is an accident of alphabetization, but I'll take it.  I do have to confess one thing...when this contest was first announced, my blog was not on it.  Fortunately, nominations were still open at that point, and with characteristic modesty, I did the honors myself.

In my category, I've got some stiff competition:  HIStalkiMedicalAppsNanoTechGalaxy, and ScienceRoll.  Click the links to check them out.  These are all pretty high-class, but none deal with the information with my level of humor, sarcasm, and self-depreciation.

Please go to the medGadget site and consider voting for Doctor Dalai.  While you're there, have a look at the other categories, too!

So far, Dalai's PACS Blog isn't doing so well at the polls. I really don't need to win the prizes, but it would be awful to come in dead last...

Friday, February 04, 2011

TATtle-tales, Paranoia, and Unintended Consequences

My favorite saying of all time is this: "Just because you're paranoid doesn't mean they're NOT out to get you." Truer words were never spoken.

We all know at some level that our PACS stores various measurements, known as physician metrics. Depending on your system, these can get pretty detailed, and may include everything from TAT (turn-around-time) to how many studies you read in whatever time period to how long you spent on any particular study. This is valuable information, if used properly. As reported in
Digital dashboarding uses data mining techniques to continuously, automatically process and visually display custom-set key performance indicators (KPIs) that provide operational data for a practices, departments or group facilities. Paul G. Nagy, PhD, associate professor of diagnostic radiology and nuclear medicine in the department of diagnostic imaging at the University of Maryland School of Medicine(UMD) in Baltimore, says dashboards not only give managers a greater holistic view of their practice, but also reduce costs and can improve productivity.
The majority of this effort seems to be geared to TAT improvement. That of course is a laudable goal, and my group does use data from the RIS to track how fast we sign off our reports once they become available. Hopefully the stopwatch starts at the time the report appears in the queue, and not when the order went in or something, because we would look a bit worse. Still, I think we have pretty good TAT, and that's without Speech Recognition, thank you.

This is all well and good. However, let's get a little paranoid and speculate about what could be done with all this mined data. Can you imagine what might happen if unfriendly types got hold of it?

"Dr. Smith, do you think the fact that you spent only 2.51 minutes reviewing my client's X-Ray had something to do with the fact that the tiny dot you missed turned into a cancer 15 years later? HMMMMM?????"

"Dr. Jones, how could you believe you were doing a good job on the day in question when your PACS system documents that you read 259 examinations without a break? Did you even have time to go to the bathroom? No?"

"Dr. Johnson, the PACS records show that you looked at my client's scan but didn't read it, leaving the life-threating finding for your partner to discover 18 hours later."

"Dr. Dalai, the logs show you were blogging when you were supposed to be helping out with the 368 examinations on the reading list at our seven hospitals."

You get the idea. Well, you say, this just keeps the damn doctors honest. Great stuff. Maybe, but hold your judgement for a moment or two.

Using physician metrics, especially as ratings, is nothing new. Our British friends at the National Health Service tried tying incentives to metrics, and, not surprisingly, the metrics improved:
In 2004, after a series of national initiatives associated with marked improvements in the quality of care, the National Health Service of the United Kingdom introduced a pay-for-performance contract for family practitioners. This contract increases existing income according to performance with respect to 146 quality indicators covering clinical care for 10 chronic diseases, organization of care, and patient experience.

English family practices attained high levels of achievement in the first year of the new pay-for-performance contract. A small number of practices appear to have achieved high scores by excluding large numbers of patients by exception reporting. More research is needed to determine whether these practices are excluding patients for sound clinical reasons or in order to increase income.
Just what we would expect. Except all may not be as it seems. Here's one analysis:

Really what they showed when they rewarded doctors for maintaining good health metrics in their patients was that doctors that treated the young, healthy, and rich did well, whereas those that served more patients, the poor, the elderly, and difficult patients were paid poorly. Also, those who filed lots of "exception reports" to justify the exclusion of a patient from the data set did the best of all.

Basically, they show that rewarding (and I suspect this will apply to penalizing as well) doctors based on patient health metrics led to doctors serving "easy" populations to do well, and those serving "difficult" populations to do poorly (or try and cook the books). There is no evidence it led to a significant improvement in care. I'm all for getting paid more, but the problem with penalizing or rewarding doctors based on how the patients perform is that it rewards doctors for avoiding difficult patients. ...

The next step will obviously be to make sure that doctors perform the cheapest procedures they can justify, and select the drugs the insurance companies prefer because of rebate deals they make with drug companies.
Here enters both the Law of Unintended Consequences and a bit of paranoia. Alan Greenspan never got it, but the rest of us do understand that people will modify their behavior in ways that benefit them. If you are going to measure something I do, I'm going to change what it is I do so that I look good on the measurement and get the prize. I can illustrate that well with our local TAT problem. As above, my group is monitored for the length of time it takes us to sign off a report. Since our Cerner RIS is a pain to sign on and use multiple times a day, many of us were a bit sluggish about getting this done. But once Big Brother, I mean our friends, started monitoring this, we found ways to improve this problem. My solution, which most of my partners have adopted, is to use a little macro program to keep the RIS window open and in fact flash it in my face every 30 minutes. Thus my TAT never gets much past one hour. Of course, this is a good result. However, with metrics that have more latitude, such as "difficult" patient populations, the practitioner is going to modify his or her approach to fit the metric.

Let's take this to the logical paranoid conclusion. "Meaningful Use" of the EMR is still pretty nebulous, as are most things the government foists upon us, but in reading through the mountains of legalese, one thing about it does become quite clear: for an EMR to be certified, it must be able to transmit data back to the Governmental Mother Ship in Washington, D.C. You can't convince me otherwise. Take the examples above and multiply them by some large fudge factor. The goal is for every doc in the country to report (theoretically anonymously...sure...) on every patient in the country. What a data-mine-field that will be! The potential for abuse is just beyond comprehension. Anyone ready to join me in my paranoia yet?

But I'll conclude on firmer ground with the PACS situation. I think it needs to be determined who owns and who can access the physician metrics stored on your friendly neighborhood PACs. Frankly, I believe this data should be non-discoverable, and available ONLY in anonymized fashion for peer-review and the like. There should be a facility for an individual to see his OWN metrics, but nothing beyond that, except perhaps for how he compares statistically to the rest of the group as a whole.

At this point, I don't know who legally owns this data, the docs, the hospital, the vendor, or the janitor. Also, I don't know if this has ever played a part in any litigation, but I guarantee you, it will someday. Because there are indeed folks out to get us.