Vaporware

Wow....long time no posts. Sorry. Something about a new baby in the house impacts the blogging frequency.

The title of this post says it all: Vaporware. For those of you who don't know, it is when a company promises you a certain upgrade in the future and then continues to push back the release date repeatedly, over and over again. Yet again, I am experiencing this from a PACS vendor (and NOT Medweb, in case you were wondering. It is working quite well.)

Our former vendor (Kodak) has been kicked to the curb earlier this month at the Level I trauma center. What an incredibly awful product. I shed no tears for it being gone. For the private equity firm that bought them: GOOD LUCK, you will need it. Polish a turd and you still have....nevermind. Note to Kodak: Autosaving image markups is not an impossible thing for other vendors.



Now for the new vendor and some background:

When it was decided to pick a new vendor, it was narrowed to 3 big name PACS companies. This was narrowed to 2 vendors in short order. Both of these vendors were quite strong. One of the key things Vendor A had was excellent remote reading capabilities with an active worklist.

Vendor B did NOT have great remote reading with an active worklist (an active worklist allows everyone logged on to tell if a study has been reviewed). Vendor B was adamant about being able to provide it by installation, and these assurances were put in writing. We went with Vendor B, needless to say. Supposedly it would be ready by RSNA...nope. Ready by install..nope. Last week: It will be out in March. Today....Well, March or April. Do you follow where this is going????

One of the guys in my group had talked to another group with experience with this vendor, he warned us about this..unfortunately it has proven to be true thus far.

The sales staff for Vendor B was all very interested in our wants and needs prior to install. Not so much anymore. Big surprise there. Maybe when they don't get paid they will wake up?????

Dalai talks about vendors and the games they play, and many of his posts are in this vein. What vendors don't realize is that complaints about them WILL come out when customers are mistreated...and it will potentially have an impact on future sales.

Who is the mystery vendor.....Well, I'll play nice this time. but if said vendor does NOT follow through on their promises by the end of March, you will see their name in bold print all over this poor excuse for a blog. It is unhealthy to keep such anger in, is it not??????

And if they do follow through....well, I'll post the appropriate kudos here as well.....

Tick...Tick....Tick..................

First Radiology Grand Rounds....

Check it out at:

http://sumerdoc.blogspot.com/2006/06/radiology-grand-rounds-volume-i.html

And now for something completely different: Kodak PACS and IDX

Well, Medweb is working well...so it's time to move on to something else. Some background:

I have been in private practice for just over a year and a half. Ever since I left residency, the major medical center that I spend a significant portion of my life @ has been using Kodak PACS. I do not consider myself a PACS expert, I simply know what works well and what doesn't. The systems I have used thus far in my short career: Agfa, Fuji, Mckesson, Stentor, and Medweb.

From a Radiologist's perspective, a PACS system should do the following:

1) Display images. Seems simple, huh? Well....not really. It should display images in a consitent manner with prior studies available for comparison. For example: It should be able to display a portable chest radiograph with the most recent portable chest radiograph for comparison. If there is not a portable chest, it should show you a PA/LAT or other relevant study. I have been waiting since I arrived @ my job for this from Kodak......and we are still being told "We are working on it". This does not even address the missing priors due to archival issues, inconsistent windowing, random display of MR sequences etc etc

2) The worklist should be active, and should let you know when someone else is looking at the same patient and/or study. This allows for a paperless department. Sorry...no way to do this either in a consitent manner with Kodak. To be fair, using IDX's worklist solution in front of Kodak doesn't work either.

3) Prior reports should be available with relevant clinical data that is pertinent to the exam.

4) The system should be stable, and not require am/pm reboots of the workstation.


My advice to the PACS vendors: Once you can do this, and do it well.....then consider proceeding to the fancy stuff: MPR's/Image Fusion etc. But don't even think about doing this UNTIL you have the above 4 items working PERFECTLY. What I see with Kodak at this juncture is "Gee...were working on getting your morning portables displaying correctly, but look, you can make MPR's!!!!" Yippee. MPR's do nothing for me when I'm rebooting and dragging and clicking 100 times a morning trying to review morning portables.

The other thing you come to realize is that some IT people (not at our hospital, but with the vendors) just don't get it. They think we, as radiologists, somehow our intimately concerned about what there database runs on etc etc. We as the end user, do not care. We are about having data available to us in an efficient manner that allows us to make appropriate findings to help our patients.

Anyway, I could ramble on forever. At our hospital, there was an attempt to integrate IDX,Kodak, Powerscribe, and Dictaphone (which I cynically refer to as dis-integration). Let's just say that it has not been a pleasurable experience, the vendors have not lived up to their promises. At this point, other alternatives are be considered. My advice in this situation: Put one company in charge, and tell them they have to make it work. Otherwise, you have one vendor blaming the other.

I have though about registering this trademark in case some PACS company want's to buy it...but I doubt there will be many takers:

"Our PACS is better than hanging film.....barely"

Thanks to our departmental IT folks....I feel for you guys.

Medweb PACS....finally, an update

Wow....Thanksgiving is here. I have many things to be thankful for, and one of them is the fact that the Medweb PACS is finally stabilized and is up and running. My last post was in September, and it was about three weeks after my post that we had the system up and running.
To summarize: Our group looked @ 3 or 4 vendors for a solution to allow us to look at images from outpatient imaging centers. Medweb came in the cheapest and included a digital transcription server. I went on a site (? Sight) visit to NYC, and we had a smaller version of their product for night call already in place. 3 or 4 other vendors were much more expensive, and frankly expected that we would have a full time IT person to run the system.

It took about four weeks longer than I expected, but we finally got the system up and running. We can now look at studies via our website virtually anywhere in the world. Radiology Dashboard works well.

What did I learn from the experience:
1) Add 2-4 weeks t your expected start time
2) Report distribution is as important as image distribution....and in many respects, is more difficult.
3) Sometimes, the only way to get problems fixed is to be a complete a#$!#ole
4) In retrospect, I would have had our part time IT person fully assist from the start in the PACS decision.

As for Medweb:

They are a small company. As I learned, this is both good and bad. They do not have huge resources, but when something needs to fixed, they can make very rapid changes. We were, in essence, a beta site for Radiology Dashboard and to some extent, the transcription server. As a result, both Medweb and my group had some headaches along the way. The upside to a small company: You can actually talk to the CEO and he can focus the company to fix critical issues.

The bottom line is Medweb is working well for us, it is a very cost effective solution, and I hope they continue to grow and improve their product.

Medweb PACS

Well....it has been a frustrating week for me...as the "Point Man" in getting the Medweb PACS and transcription server up and running. Our goal was switching to it on Thursday....looks like it won't be for 1-2 weeks...do to a multitude of factors. The events:

1) No transcriptionist...despite hounding Medweb for 4 weeks and them knowing that since September 2nd that our transcriptionist couldn's use her footpedal...she still isn't up and runing. The apparent solution: Buying her a $60 USB footpedal compatible with their product. Unbelievable....

2) Our network guys: Talked us into buying a $5000 Cisco firewall and router for our old sytem claiming better speed and throughput..which never materialized. They then claimed they chose it because it would work well with any future PACS solution. Then...they couldn't get it to work....Then they contacted CISCO..Cisco's reply (and I thought Medicine had it's own language):

First off, the static inside, outside resides on the outside of the asa,
> so
> even if you try to speak to that address the asa does not allow you to
> speak
> to an opposite interface of the device. So if you're on the inside you
> can
> speak to devices on the outside as long as the addresses for them are not
> maintained by the ASA, but you cannot actually speak to anything that is
> in
> reality the outside interface of the ASA which static addresses are.
>
> If the application only needs to initiate traffic to it's public address
> and
> is capable of doing so by it's hostname/domainname then you may be able to
> get away with doing the DNS fixup to rewrite the dns response so it gets
> it's internal address.




The network guys then decided to do away with the VPN..not a good idea. Medweb is apparently going to congfigure the box for us Monday with the VPN set up between our imagng centers and the firewall to their two servers. Our network guys are very well paid to plug and unplug ethernet cables.

3) The poor guy Medweb sent us really didn't seem to understand the workflow between the transcription server and radiologists..who chewed him up and spit him out during an evening training session. Nice guy from Medweb....Medweb should be ashamed for sending out someone w/o adequate training.

4) On a positive note: We are one of the first sites using Medweb's "Radiology Dashboard". It downloads studies automatically when you log on in the background and has a lot of neat features. HOWEVER...on Tuesday we realized it had two major problems: The new version wouldn't work with efilm...and the USB transcription microphone buttons wouldn't work. Two major issues....which Medweb fixed overnight.

The bottom line: The verdict is still out...look for futrher updates here!

Case 3 of 3

Brown Fat: I saw 3 cases today…it was freezing @ the PET center. Brown fat represents the attempt by the body to stay warm by decoupling the formation of ATP to directly produce heat (I'm NOT a biochemist). It is more commonly seen in the winter…and often times if the air conditioner is running full blast, you will notice multiple patients on the same day having this pattern. It is typically in the neck and paraspinal regions…the fusion images help tremendously. Valium administration helps decrease this…and may be needed in Head and Neck Cancer patients.

Case 2 of 3

This patient had 2 PET negative pulmonary nodules, both about this size. False negative nodules are seen typically with mucinous adenocaricinoma (particularly well differentiated), carcinoid, and bronchioalveloar cell.

The patient had mucinous adenocarcinoma of the colon, I suspect these are metastatic lesions despite the PET findings.

3 PET Cases

Breast Cancer Patient: Turns out this was bone marrow hyperplasia. You can see this with granulocyte colony stimulating factor, severe anemia, and rebound after chemotherapy. Hints that this is the dx is how remarkably homogeneous the uptake is (mets are typically more “mottled”), and the fact the patient had a negative bone scan.

Expert Witness: In a perfect world

Dalai has a post that talks about being sued...something that hasn't happened to me...but will. I am recently out of training...which is probably why I haven't been sued yet.

Expert witnessess, as Dalai points out, are in many instances nothing more than whores (my wording, not his) to state whatever their side wants them to.

What I would like to see (and you could do this in radiology...it is a digital medium) is the ACR or some other professional society sponsor a case review panel.

Example: You are sued for calling a lesion a calcified granuloma. As it turns out, it was cancer, and the "central calcification" is two pixels off when you magnify it on the PACS monitor to the subatomic level. This case could be sent to all the experts electronically for review. They could then render their opinion. Then, in court, the jury could be told that 99 out of 100 radiologists said they would have called it the same way.....this would carry far more weight then 2 dueling expert witnessess.

To me...this makes perfect sense....which is exactly why it won't happen.

By the way, the above example came from:

THE PRACTICE OF RADIOLOGY:
Leonard Berlin
Failure to Diagnose Lung Cancer: Anatomy of a Malpractice Trial
Am. J. Roentgenol., Jan 2003; 180: 37 - 45.

A VERY depressing article!

PACS Update

Our group will soon be installing our PACS system: http://www.medweb.com/

Many bridges will need to be crossed:
1) Installing a Sun image server and a transcription server
2) Training the transcriptionists on the new server
3) Getting reports to integrate into the various sites IT systems...a real headache for one site
4) Getting the various imaging centers connected to the new image server

The goal of our changeover:
1) Enable us to review studies anywhere, anytime with internet access and be able to render prelims in a digital manner, both with audio and text.

I'll let you know how it goes....and would love to hear from anyone who has gone through this!!

Thyroid Therapies.....what do you do???

The Scenario:

The patient shows up and you are told that Dr....... (the patients Endocrinologist) has written for them to be treated with XXX Mci of I131 (With XXX being way too much or way too little, in your honest opinion). Do you:

1) Say ok, that would be great, not talk to the patient sign the script and dictate it later?

2) Do you become offended, call up the Doc in question and have a good old fashioned knock-down-drag-out argument?

3) Do you call up the Doc and have a try to have a collegial discussion about what would be the best thing for the patient (with variable success)?

I've been in Radiology for years, both as a Radiologist and a former technologist, and I've seen a tremendous variability in the how the above scenario is addressed. My personal approach is to try to talk to the Endocrinologist in question and try to explain my reasoning etc. Quite honestly, I think they would rather just have me shut up and dose the patient. I wonder, at least to some degree, if this isn't a side effect of an overall lack of thyroid therapy interest on the part of Radiologists, as well as the perception that Radiologists don't understand the pathology of hyperthyroidism......

If there are any Endocrinologists out there...I'd be interested in what you think.

On another note....if anyone has comments that they'd rather not post, I can be emailed at: nuclearvision@gmail.com...it will forward to my real email account.

Importance of Blood Glucose and PET

The above case shows a node in the lt illiac region. The node on Monday's
scan is barely positive with an SUV of 3.1 (lower image) , the patient took their full dose
of am insulin, ate, and had a glucose of 145. There is a marked amount of
muscular activity from residual SQ insulin. I had him come back 2 days later
after not eating and only taking 1/2 his am dose, the glucose level was in
the 60's on the repeat.

SUV on repeat: 10.3

Makes you wonder what the node would look like with a glucose of 200.

PACS...Choosing a vendor

After much debate, our group has decided on a PACS vendor. A description of our needs:

The ability to retrieve images from a central server to any site in the world, but primarily to 5 Dell workstations in the local area where we work. Digital dictation is a must. Currently, we have 3 outpatient sites sending to a central server (1 site is CT only, the other 2 our MRI only) that then allows retrieval by 4 other workstations via T1 lines. Current data sent to our server/day=2.5 gigabytes (30-50 studies). The problem with our current server is it is essentially unable to function at our remote locations. Our group contracted with 2 local companies, one network oriented and one software oriented.

We like both of them on a personal level. However, they blame the current lack of functionality on each other. The network people tend to blame the software people and vice versa.

We asked for quotes from Stentor, Amicas, Agfa, and Medweb.

Stentor and Agfa came in very high. They aren't targeted at a group our size...and these bigger companies don't understand that we don't have a PACS administrator etc.

Amicas and Medweb came in close together, we have a small Medweb server for night time stuff and have been reasonably pleased with it. They also have a transcription server which we are getting. We are also going with a co-location server..which should improve our bandwidth.

Go live date is 9-15-05....I'll keep you updated and would appreciate any input.

Aunt Minnie

Diagnosis???



































Hypercalcemia. Patient had a bone scan done 4 days prior that was identical, the tech's thought there was free Tc99m present or that the patient had another study at an outside institution....they didn't consult either myself or my partners and decided to try again...with the same results.

Welcome

New to the blog world....hope to put my musings about radiology and nuclear medicine on here....w!!ish me luck