Sunday, March 26, 2017

The Best Designs Of (For) Mice And Men

Of all the things you didn't know about me, perhaps the most irrelevant is the fact that I'm a frustrated inventor. Periodically, I come up with ideas of things that should exist but don't, and in general they don't for some very good reason. I did once go so far as to retain a patent attorney to research one of my brilliant intellectual offspring. He found the same thing had been "discovered" and patented four times before. That knowledge cost me $500. Of course, one would search the U.S. Patent Database today with Google. How times have changed.

The Agfa Daily Blog Update recently linked to an article celebrating some of the best designed products of all time.  It's a good read, although I found the selection a little, well, optimistic for that particular blog. Still, it prompts some interesting discussion. Five design professors were interviewed, and asked for their choice of the best-designed product of all time. I won't go through the entire list, but the one that took my fancy was the lowly old dial telephone, as described by Professor Kalle Lyytinen of Case Western Reserve University:

American industrial designer Henry Dreyfuss’ AT&T Model 500 phone is one of the most iconic and recognizable products of the 20th century. The phone – together with its design process – was a harbinger of many design principles used today.

Rotary phones – which feature a round dial with finger holes – first emerged in the early 20th century. But many of these were bolted to the walls or required two separate devices for speaking and listening.

In addition, early telephone users would call into operators, who would use a switchboard to connect callers. When this process became automated, designers needed to figure out a way to offer an intuitive interface, since callers would be dialing more complicated number sequences (essentially doing the “switching” on their own).

Though earlier models came close to addressing these needs, the 500 model elevated the design, adding several functions that forever changed the way phones would be used.

AT&T’s first rotary phone in 1927 (dubbed “the French Phone”) had an integrated handset for both the loudspeaker and the microphone, but it was cumbersome to use. Meanwhile, Dreyfuss’ earlier model from 1936, the 302, was made out of metal and also had an awkwardly shaped handset.

Then, in 1949, his Model 500 came along.

Employing new plastic technology, the phone’s handset was smooth, rounded and proportional, an improvement on unwieldy earlier versions. It was the first to use letters below the numbers in the rotary – a boon for businesses, since phone numbers could now be advertised (and remembered) as mnemonic phrases (think American Express’ “1-800-THE-CARD”).

The 500 phone was also the first to undergo a design process that used ergonomic (comfort) and cognitive experts. AT&T and Dreyfuss hired John Karlin, the first industrial psychologist in the world, to conduct experiments to evaluate aspects of the design. Through extensive consumer testing, the designers were able to tweak all minutiae of the product – even minor details like placing white dots beneath the holes in the finger wheel and the length of the cord.

Including its later incarnations, the phone would go on to sell nearly 162 million units – around one per American household – and become a presence in living rooms, kitchens and offices for decades to come.
Italics mine.

It should be intuitive that a well-designed product does what it is supposed to do, does it well, and is easy to use. Compared to this old phone, the 500 was quite an advancement in each of those criteria:

As I've said on numerous occasions, Apple (whose products were curiously not mentioned among the top five) has mastered this concept. The late, great Steve Jobs quite literally used Zen philosophy in his product design. As Drake Baer writes in Business Insider:

When you look back at Jobs' career, it's easy to spot the influence of Zen. For 1300 years, Zen has instilled in its practitioners a commitment to courage, resoluteness, and austerity — as well as rigorous simplicity.

Or, to put it into Apple argot, insane simplicity.

Zen is everywhere in the company's design...

It's the industrial design equivalent of the enso, or hand-drawn circle, the most fundamental form of Zen visual art.

But Zen didn't just inform the aesthetic that Jobs had an intense commitment to, it shaped the way he understood his customers. He famously said that his task wasn't to give people what they said they wanted; it was to give them what they didn't know they needed.

"Instead of relying on market research, [Jobs] honed his version of empathy — an intimate intuition about the desires of his customers," Isaacson said.
It is rather ironic that Agfa itself attempted to develop a PACS interface in a vaguely similar manner using Alan Cooper's Persona approach with limited success, depending upon whom you ask. We still use IMPAX 6.x, which is the one of the later descendant of Agfa and Cooper's Odyssey PACS prototype. It does work, but takes approaches I would not, as a practicing radiologist, have recommended, and I continue to despise. As usual, the bottom line is this: those who design products MUST get into the heads of those who will USE those products. It really is that simple. Steve Jobs got it. Tim Cook, maybe not so much. Some PACS companies, well, not much. Maybe not at all.

I've been blogging about PACS for over 12 years now, and I'm not seeing a whole lot of improvement in this regard. Here's a good example. I've spoken previously about our Centricity Universal Viewer, which is not universal in any real sense, although as the heir to Dynamic Imaging's IntegradWeb, it had great potential. We have been able to come to terms with it, and GE has actually fixed many of the problems we've had with it. But as my senior-most partner puts it, the enemy of good is better. Exhibiting the faith of those who walk on hot coals and handle snakes, we agreed to have the embedded version of the Advantage Workstation placed on the system. It seemed like a good idea...we would be able to view PET/CT's and do high-level imaging things on any workstation. The number-crunching is done on a separate AW server with server-side rendering, so there should be no ill-effect upon PACS. Right.

In practice, well, we've had some trouble. The integration of these two VERY different products could not have been easy, but it could have been done better. As usual, it appears that no radiologists were killed in the making of this product. Or consulted, for that matter. And, those at GE who know the UV well don't have expertise in the eAW (embedded Advantage Workstation) and vice versa. So it is no surprise that the integration of the two is not what it should be. Without going into painful detail, say we are viewing a PET/CT with a comparison conventional CT. I'll have the CT images from both arranged on the left side of the 6MP monitor, and the server-side rendered (but still SLOW) AW windows the right. As originally configured, scrolling through the CT was supposed to synchronize all windows. But that ended up moving images around in an uncontrollable way. I asked for this connection to be severed, and GE tried to do so, but some crosstalk does remain. For example, changing a CT window on UV changes it on eAW as well. Trying to load a different comparison CT restarts the eAW window altogether. And so on.

Had I been called upon to choreograph this dance, I might have been tempted to do some of the synchronization, but I would have limited the depth of the connection. It needs to be kept simple, in my humble (and simplistic) opinion. One side really must not be allowed to interfere with the other. I should think it would have been easier to make the windows totally separate in their operation, so what we see here is a perversion of Job's philosophy. We are given something we didn't know we wanted, and lo and behold, we really don't want it after all!  There are additional problems with hanging protocols, which are completely different entities on the UV and the eAW, but may create overlap. GE has been helpful, but I have the feeling they have not encountered these problems before. Perhaps we are doing something very wrong, or maybe this is one of the first installations of this particular patois of hardware and software. We await further tweaking.

I'll keep you posted.

I was once asked if I planned to create my own PACS. For better or worse, I don't have the resources, the backing, or the expertise to try this, but I am available for consultation (for a very reasonable fee) should anyone with a lot of money and a team of software engineers be interested in making The Best PACS. I'll be waiting by the phone. If contracted for this lofty purpose, I'll certainly do my best to channel the spirit of Steve Jobs. I can try, anyway...

Post Script: My friend Phil sent this comment concerning PACS development, and it is certainly worthy of inclusion into the main post:
When will they ever learn, oh when will they every learn?

Answer - When it affects the bottom line.

Question: When will that happen?

Answer - When radiologists are empowered to decide which product to buy (Oh - and will spend the time to actually evaluate them).

Question: When will that happen?

Answer: ????????????

Sunday, March 19, 2017

Hello, (Friends of Doctor) Dolly!

I know many of you are landing here thanks to my daughter, Dr. Dolly. She was just published on, and my celebratory post on Facebook labeled her a "chip off the old Doctor Dalai". Thus, her friends are now discovering what I do in my spare time. Hopefully this won't reflect badly on Dolly. She is, after all, at the beginning of her career, and I'm at the end of mine. We don't want potential employers, colleagues, administrators, scrub-nurses, or Uber-drivers to think she might turn out like me! (For privacy reasons, I'm not linking back to KevinMD.)

In some ways, I've taken a page from her book. While in medical school, Dolly went on a number of mission trips to such amazing places as Oaxaca (Mexico), Nicaragua, and Uganda. And South Dakota. Having more time available and more training behind me, it occurred to me to do the same while bouncing around the purgatory of quasi-retirement. Once I found RAD-AID, the die was cast. Both of my loyal readers know that I've been to Korle Bu Teaching Hospital in Accra, Ghana, an incredible trip. You can read about it right here on my blog if you haven't already. A medical mission trip is not something you do once; the experience changes you (for the better). The desire to give back, and the growth involved in the process, is addictive. The friends you make, the things you see, the joy of being out of your comfort-zone all necessarily call for an encore performance.

Thanks to a tremendous opportunity provided by RAD-AID and SNMMI, I will be going to Tanzania this summer to provide what little expertise I can muster for the Nuclear Medicine program at the Aga Khan Hospital in Dar es Salaam.

The whole thing comes as a bit of a surprise to me, as I will be the recipient of a Hyman-Ghesani RAD-AID SNMMI Global Health Scholarship, which will cover much of the expense of the trip. The surprise is that this program seemed to be geared more toward academia, and I applied with little hope of success. But I seemed to have impressed the committee to an adequate degree and so off to Tanzania we go. I am truly honored and humbled by the confidence and trust in me. I do have to say that in my 27 years of private practice, I've come to find that experience is the best teacher. Of course, experience and brilliance would have been a better combination, but we can't have everything.

The mission has only one downside. I'm committed to present a report at the subsequent Society of Nuclear Medicine and Molecular Imaging meeting, and in 2018, it's in...Philadelphia. Meh. Oh, well, we have to make some sacrifices here and there.

I've said it before, and I'll repeat...RAD-AID is an incredible operation, worthy of your donations of money, and better, of your time. There is tremendous need out there for your radiological expertise, and yes, your cash. There is a lot of work to be done. Come join me.

Saturday, February 18, 2017

Out Of Antarctica

I had intended to post from Antarctica itself, but time somehow gets away from you wile in a place like this. So, I'm posting instead from the Scotia Sea en route to the Falklands.

You might have heard of the Drake Passage between Ushuaia and the Antarctic. It is often windy and treacherous, and has become known to those with weak stomachs (such as yours truly) as the Drake Shake. Fortunately, we had one of the quieter rides, and so we refer to this area alternatively as the Drake Lake. (We've had some rough seas here and there anyway, and I'm trying to keep my meals from repatriating themselves to the outside world with an Australian drug called TravaCalm. So far, so good.) And believe it or not, there were a number of other cruise and expedition ships down there with us. Antarctica is becoming a major eco-tourist attraction.

Our week in the REALLY Deep South included four trips out to land, twice to islands, and twice to the mainland peninsula itself. Given the latter, I have officially joined the 7 Continents club, having now set foot on all seven continents. (Naturally, some academic types have just now come up with a possible eighth continent, Zealandia, but it's mostly under the ocean, so in my book, it doesn't count.) There were two days with weather rough enough to keep us away from shore, so we got to steam around and see the incredible territory instead. That's a reasonable consolation prize.

Photos do not do this area justice. At all. At least mine don't. Of course, my photo equipment includes a Sony RX100 M3, a GoPro with stabilizer gimbal grip, and my iPhone. I'm thinking the iPhone is probably the best of the lot. I've had some terrible cases of lens-envy when observing the setups some of the other passengers have with them. I attempted to ask someone who was showing off a few really incredible whale pics about his set-up. "It's a Canon 5D Mark you know cameras?" When I said I did know something about them, but was not in possession of this $5K setup, the gentleman then made it a point to ignore me. See my comments about class in the previous entry.

But I do have hundreds of photos, and I'm proud of them. When I get back home, and have something to process them beyond Mrs. Dalai's old MacBook Air, I'll try to compile the best of them. In the meantime, here are a few random pics:

You'll notice white lines in some of the images of the penguins. I don't have to tell you what they are, do I? But where there's grant money, there will be someone to claim it. The image below is from an honest-to-gosh scientific paper about...penguin defecation:

Yes, this was the topic of an on-board lecture. No guano. 

And finally, it wouldn't be Valentine's Day in Antarctica without a dip in the hot tub! It wasn't too bad getting in...

We were to get one last penguin visit in the Falklands, but gale-force winds forced us to carry on toward Uruguay. So w are now officially in penguin withdrawal, if there is such a thing. Oh must be flexible when traveling in this part of the world.

Friday, February 10, 2017

Politicians and Penguins

My level of love for animals is mostly restricted to our two little fluff-balls at home, whose vet and kibble bills top the GDP of several small nations. Mrs. Dalai, however, LOVES animals, especially penguins. And so, when deciding where to go for our next big trip, she was quite clear on where she wanted to go. Antarctica!

When Mrs. Dalai speaks, I listen. And so, we are presently on a ship traversing the Drake Passage, and will indeed reach Antarctica by tomorrow morning.

Yes, it's getting colder...Supposedly the temperatures on the Antarctic Peninsula itself will be in the low 30's Fahrenheit, not quite as bad as I expected. 

We started the cruise in Valparaiso, Chile, heading down the west coast of South America, stopping at various ports in Chile, and then Ushuaia, Argentina, the southern-most city in the world. 

From Puerto Montt, we bussed to the Pertohue Falls:

On the island of Chiloe', we visited a number of wooden churches, designated UNESCO World Heritage sites, and these palafitos, houses on stilts, in the city of Castro.  

From Punta Arenas, we took a speedboat to Magdalena Island, and finally saw penguins! There are a number of penguin species, and these are Magellenic, also known as Jack Ass Penguins. 

There are a number of glaciers this far south, including this one in the Chilean Fjords:

And tomorrow, we reach our first adventure stop, Half Moon Island, in the South Shetlands just off the coast of Antarctica. We'll be here for a week, with an hour out on the ice each day. We've brought enough winter gear, including hand and foot warmers, to last a lifetime. Weather-wise, at least, I prefer my version of the South to this version of the South, but I must admit that the scenery here is much more dramatic.

You might wonder who would undertake such a trip; I certainly did. As you would guess, the passengers are by and large older, almost all are older than we are. Who else is going to take almost 4 weeks to make this trip? (The cruise goes on to the Brazilian Amazon and then to Florida for those who can take two MONTHS off.)  The average is somewhere south of dead, but on its last run through this area, the ship had to leave the Antarctic on an emergency basis as two passengers took ill. There is NO way to evacuate someone whilst here. One of the two did not survive, I'm told, but was content to leave this Earth doing what he loved best. If I ran the cruise line, I would require medical certification before ever allowing anyone on this sort of cruise, but that's just me.

Part of the fun of this trip, however, is indeed the folks we meet. We've come to know the retired CEO of a large aircraft company, several self-made men (and women), the retired head of the Social Security equivalent of a major Scandinavian nation, and while I've not tried to disturb his privacy, a former presidential candidate and his perfectly-coiffured wife are on the trip as well. And there are the usual complement of folks we try to avoid. "You're from the SOUTH? We don't know anyone from the SOUTH. They don't think like WE do." "Morris!!!!!! They're all out of cucumbers at the salad bar! DO SOMETHING!" "We only get a few weeks off. Not like Jewish people who always have extra holidays..." Money doesn't buy class, it seems...

Anyway, we are on the first group to hit the land tomorrow, so off to bed. Stay tuned for the Penguin Report, coming to these pages very soon.

Tuesday, January 17, 2017

Fovia Utopia

I've found over the years that the PACS field is rather mobile. A friend that worked for company A might now be found at company C having arrived there after doing time at firm B. You might remember back when Merge first bought out Amicas...there were some discussions about whom to keep and whom to let go, and I was rather vocal in urging (begging?) Merge to keep the folks who made Amicas Merge PACS what it is. That didn't work out, but one of my friends from that realm is now with an amazing little company called Fovia. I sought him out at RSNA, as I usually do, and he showed me some cool stuff. Due to some personal things, I've been too lazy to write postponed my brief but exciting RSNA report from Fovia, but here it is. Finally.

Fovia is in some ways the biggest PACS company you've never heard of. Although it isn't a PACS company, and you have heard of it here on these very pages.

Fovia is a "behind the scenes" company. Think Intel. They make the chips, but you can't buy an Intel-branded computer. Fovia makes rendering software used by many PACS companies, including Merge. In case you didn't know, Fovia:
  • Founded in 2003 and inventor of High Definition Volume Rendering®
  • Provides a highly scalable CPU-based, server-side rendering solution delivering images to clients applications anytime and anywhere
  • Optimized for the CPU and delivers near perfect linear scalability as the # of cores increases. Does not require a GPU, or the headaches associated with OpenGL drivers.
  • Scales well with dataset size. There is almost no visible difference in performance or quality when rendering 500 slices or 4000 slices (maybe 10% slowdown)
  • Scales well and preserves the quality and performance when rendering to large displays. This is critical for larger medical monitors such as 4K or 6+ MP displays.
  • Provides a “Fovia Inside” solution that is directly integrated into OEM customers’ solutions. The benefit is that OEM customers retain their own user interface and workflows, but gain HDVR inside their products. 
  • Fovia products are integrated into products and PACS from quite a few names you know, including GE, Sectra, Merge/IBM, and Intelerad.
 At RSNA, Fovia introduced us to their two latest technologies: XStream® HDVR® WebSDK and F.A.S.T.® Interactive Segmentation.


An SDK is a Software Development Kit, and this one delivers Fovia’s High Definition Volume Rendering from the enterprise or cloud over cellular and Wi-Fi networks directly to desktop or mobile devices with no compromise in performance or quality. This yields a zero-footprint application that will run on almost any platform, be it desktop, tablet, iOS, Android, what-have-you. But this is far from just another run-of-the-mill basic viewer. With HDVR, one has access to fully interactive 2D, MPR, and even 3D workflows. The software vendor can use these building blocks to create workflows within their own products.

Click below to experience, first-hand, the incredible rendering power available from the cloud:

F.A.S.T. Interactive Segmentation

Three-Dimensional images are pretty, and often useful on their own. But the real magic comes when you can pick some particular part of the rendering to play with, which we call segmentation. F.A.S.T. Interactive Segmentation (and I'm not sure what the acronym stands for, but the darn thing is indeed fast!) F.A.S.T. Interactive Segmentation provides real-time (yup, that fast) 3D interactive segmentation on 2D and 3D views.

Now we've all seen segmentation with the various advanced visualization programs out there. But F.A.S.T. Interactive Segmentation is different: it is intuitive and easy to use. One simply clicks and drags, be it with a finger or mouse, and voila! Your favorite body part is selected and highlighted in the color of your choice. This process can work independently or can be used to augment an automatic segmentation algorithm. F.A.S.T. Interactive Segmentation is ideal for the challenging or non-perfect scans where it is critical that the case can be segmented for analysis of the disease.

With F.A.S.T. Interactive Segmentation, segmentation workflows can be completed interactively in well under a minute on a variety of studies, and not the “perfect case.” I saw the following cases demonstrated live. The screen captures don't do the process justice, but you can easily see the quality of the end result.

Low-dose lung screening using MIP or Faded MIP, can quickly identify lung nodules, segment the nodules and determine volumetric measurements (this literally takes just a few seconds). Here, the trachea and the pulmonary vasculature are segmented to determine if this lesion could be accessed via transbronchial biopsy (it could NOT)...

Here, is an example of surgical planning for a partial nephrectomy. Typically, segmentation might take between 30 and 45 minutes per case, but using F.A.S.T. Interactive Segmentation (and no pre-processing), the Fovia folks were able to do this in under 3 minutes:

  • segment the aorta and the renal arteries
  • segment the lesions
  • segment the kidneys
  • segment the renal veins

    The images are impressive, but actually watching this process take place is staggering, particularly to those of us who use advanced visualization products regularly. Frankly, I was so impressed, I declared, "Why aren’t others, like TeraRecon using this???" It seems that Fovia and TeraRecon have been unable to connect. So, believing myself to be far more important in this space than I really am, I grabbed my iPhone and called our TeraRecon rep...who happened NOT to be at RSNA. But he made some calls and we all sent texts...and nothing happened. At least not yet. As usual, my influence is less than I would like it to be, and far less than some of you out there might think.

    But I still hope TeraRecon will get around to talking with Fovia about F.A.S.T. Interactive Segmentation. Trust me, it would significantly improve the TR segmentation interface.

    Frankly, in my humble opinion, it wouldn't be such a bad idea for Fovia to offer its own PACS or advanced visualization product. Don't tell anyone I said so...

    Thursday, December 15, 2016

    Don't Spamalot
    ...Or Even A Little!!!

    We all know what SPAM is...unsolicited e-mail that clogs your inbox. But as with beauty, the definition of SPAM may well be in the eyes of the beholder. Or the SPAMMER.

    After receiving yet another piece of shi... I mean SPAM from someone who friended me on LinkedIn for the express purpose of SPAMMING, I wrote this little article for consumption on that site:
    A follow up from last year's SPAM post. DO NOT USE LinkedIn to SPAM other members. I've made the mistake of accepting contact requests, and my new "friend" proceeds to bombard me with messages and emails about their "wonderful new product/software/service/doodad/widget that I would really be interested in and would appreciate the time to contact you or whoever in your organization makes such decisions so I can share this wonderful new development...." Sound familiar?

    To all you salespeople out there...DON'T MAKE THIS MISTAKE AGAIN! Cold calls, unsolicited emails, etc., etc., accomplish nothing more than pissing people off. NOTHING. We will not be buying your incredible product, but we WILL be reporting you to your boss, your ISP, LinkedIn, Facebook, or wherever your unwanted communication came from. I have never, ever made a purchase based on a cold-call or cold email, and I NEVER will.

    I suspect I speak for quite a few of us out there who have been the targets of your unwanted missives. Find a different approach. Or a different business.
    Needless to say, a couple of salespeople were not amused. I'll keep their names and companies private, but these were quite available on LinkedIn...

    A gentleman based in a subcontinent on a different side of the world asked this in response:
    Then how do you want a sales guy to approach what is your thought for a sales guy.
    To which I responded...
    I was waiting on someone to ask that. You MIGHT find someone who appreciates cold emails. I don't, and I don't know anyone who does. Getting one of these unsolicited emails guarantees that I will NEVER look at what you have. There is almost always a dead giveaway wherein your colleagues ask that I "forward this to the person in my organization who would be interested/in charge." That will NEVER, EVER happen. And getting names from a list you purchased is one sure way to alienate me forever. DON'T DO IT!!!! Frankly, I and just about everyone I know do NOT want any unsolicited email from sales people at all. EVER. IF your company has the next best thing, have your CEO or CTO contact me. BUT NO SALES PEOPLE. Your colleagues have done a very good job of burning that bridge.
    A Sales Manager for a small IT company then wrote a rather scathing response, augmented by the fact that he once worked for a company I befriended. No names. He went off on a bit of a tear, agreeing that perhaps LinkedIn shouldn't be used for SPAMMING, but then expressed his great distress (perhaps not quite as nicely as I did) that I was casting salespeople as deplorables and trying to take food from the mouths of the salespeople's children. Just call me Dr. Scrooge, I guess. Mr. Manager went on to suggest, from knowledge acquired in his prior life, that I had had problems with ER docs listening to me which was somehow supposed to be analogous to receiving cold sales-calls. The other gentleman from overseas joined in, saying that, "customers are mean."

    You can guess how well that went over with me...
    Mr. Sales Manager, you might seriously want to remove that comment. Your seniors at (your company) as well as all of your LinkedIn contacts just saw your rant and your less-than savory approach to a friend of a place you used to work. Rather bad form. Same for Mr. Overseas. It's a really bad idea to call customers "mean". Your analogy is faulty, by the way. I have a relationship with the ER docs. Completely separate issue. I have NO relationship with Mr. Overseas and all the others (often from overseas, btw) who get my name off a list of emails they bought from some unsavory operation and proceed to send a barrage of unsolicited emails. It is the salespeople who participate in these bottom-feeder behaviors that have spoiled things for the rest of you. Try policing your own before getting angry and biting the hand that might feed you.
    Mr. Manager yanked his post, for which I congratulated him, but even then, he doubled down...
    And I think you should delete yours as well as it is still offensive.
    My final answer:
    You might want to actually address my complaint about sales people rather than digging in your heels and creating an even deeper hole to climb out of. No, I will not be removing this comment. Perhaps you need to read it again.
    And there it stands until someone else jumps in.

    I really don't like conversations of this sort, but I won't shy away from them. I absolutely, positively WILL NOT respond to a cold-call or a cold SPAM e-mail. I view this as intrusive, as a sign of desperation by the sales people involved, which tells me automatically that their product is of considerably less stature than their rosy, scintillating prose would have me believe. I cannot believe any sales person actually thinks cold-calls of this sort will actually generate any business.

    Maybe there's another approach. Let's create a website for these companies to show their wares. Then those who are interested could check in periodically to see what's new in some particular category. And communications could progress from there. We could call the site or maybe Just don't send me any SPAM to advertise it!

    Let me throw the question out to the audience...How do YOU feel about SPAM and cold-calls from salespeople who have found your name on some list they purchased for $.02 per click? Do you appreciate the warm, human contact? Or would you rather they leave you the heck alone? I don't think I have to tell you how I feel, but maybe I'm just a cantankerous old fool whom time and technology have passed by... So please do comment and share your opinion. Will it be "Sale of the Century", or "Death of a Salesman"? wants to know!

    Sunday, December 11, 2016

    Artificial Intelligence at RSNA:
    I'm Sorry, Dave. I'm Afraid I Won't Be Taking Over...

    After a pleasant Thanksgiving with the entire family, and a quick turn-around to Chicago, I had the pleasure of my 20th (I think) RSNA. I've likened this Meeting of Meetings in the past to taking a drink from a fire-hose, and that description stands. But as I get older, the meeting takes more out of me, and at times, I probably look a bit shell-shocked:

    I tried and tried to come up with a story to pervert parody to submit in place of the PACSMan awards, since my friend Mike Cannavo is otherwise occupied this year. The "Beyond Imaging" theme offered so many different directions, I found myself with a bad case of writer's block, and so no story ensued. To me, "Beyond Imaging" is that same old tired meme that tells us to act more like clinicians so patients will love us and we'll get more stuff from the ACO's. Well, maybe ACO's are going away someday, so the joke may be on us. Again. Besides, I would be rather miffed if I found that the new motto for the American College of Internal Medicine was "Beyond Clinical Medicine" implying that they were going to act more like radiologists.

    I made the rounds, attended various lectures, mainly PET/CT and SPECT/CT talks, which were quite informative, and collected an adequate amount of CME's to justify the stratospheric cost of a last-minute plane ticket, and a corner room at the McCormick Hyatt. Perhaps the most important lesson was that if you want a diagnostic CT image from your SPECT/CT, it should probably have a diagnostic CT component attached to it as the machine that produced this image doesn't:

    I spent a good bit of time taking with and about Rad-Aid, and even spent an hour or so behind the desk at their booth:

    Everyone who stopped by got the official Doctor Dalai Business Card, and a very enthusiastic retelling of my adventures in Ghana. In all seriousness, there seems to be a LOT of interest in giving back in this manner, and I could not be more thrilled and honored to be a part of it.

    I spent only a few moments talking with my friends at Merge discussing PACS. Having had to learn version 7.x on my own whilst in Ghana, I can tell you it has some new features, such as worklists built with block structures, a novel approach. It took me some time to get used to the new back end, which now divides properties among two different management areas. With greater power comes greater complexity...

    Of course, the BIG DISCUSSION all over RSNA was Artificial Intelligence, and in particular, AI as applied to Radiology. Well, let's be even more specific. There was a cloud (pun intended) hanging over McCormick, the specter of RSNA Yet To Come, which I quite presciently predicted in my 2011 RSNA Christmas Carol:
    I sat down on a PET/CT gantry and bowed my head. The room spun, and when I looked up again, we were seated on a bench beside Lake Michigan. It was a blustery day, with winds one only sees in Chicago in the winter. Strangely, I felt no chill, as I watched leaves blowing through the PACSman's shadowy figure.

    I looked behind me and gasped. The once-stately Lakeside Center was in ruins, shattered black pillars and glass everywhere.

    "PACSman! What happened here?"

    "Oy, Dalai, you need to lay off the Kung Pao, OK? Welcome to RSNA 2045," he said. "Or, well, it would have been if there still was an RSNA. Which there isn't."

    "But why?"

    "What did you expect?" he said. "Between the UnAffordable Care Act, the doctors' 'fix' that fixed you guys good, and all of your good friends, the clinicians, you radiologists didn't stand a chance."

    "But who reads imaging studies now?" I asked.

    "Geez, Dalai, why do you even care? OK, OK," he said. "You've come this far. Look, imaging reached the point where it didn't pay squat, right? So no one wanted to do it anymore. Even physicians' assistants and nurse practitioners wouldn't touch it. Imaging got so cheap that people got their scans at Walmart and everybody's data were stored in the cloud or on some vulture -- I mean, vendor-neutral -- archive. Got that? So many images were crammed into all these interconnecting networks that ... badda bing, badda boom, they grew self-aware. So, the damn computers are doing the diagnosing themselves. Whaddya think of that? End of the line for radiology."

    "No, PACSman!" I exclaimed. "It cannot be! This is an honorable profession, and it cannot end this way!"
    I would love to take credit for the current hysteria, which would mean that vast numbers of you out there actually read my stuff, which we all know is not the case. No, my colleagues have manifested this paranoia without my help. The demise of Radiology has been predicted for years, in various forms, from numerous causes, and with timelines anywhere from yesterday to 100 years from now. The latest incarnation of this sooth-saying comes from none other than Ezekiel Emanuel,  the physician brother of Hizzoner Rahm Emanuel, Boss Mayor of Chicago. Ezekiel has had his hand in a lot of, shall we say, progressive medical policies, and I think it's not unreasonable to say that he hates and/or despises other physicians. So it comes as no surprise that he and colleagues write in "Predicting the Future — Big Data, Machine Learning, and Clinical Medicine," in the New England Journal of (Esoteric) Medicine:
    (M)achine learning will displace much of the work of radiologists and anatomical pathologists. These physicians focus largely on interpreting digitized images, which can easily be fed directly to algorithms instead. Massive imaging data sets, combined with recent advances in computer vision, will drive rapid improvements in performance, and machine accuracy will soon exceed that of humans. Indeed, radiology is already part-way there: algorithms can replace a second radiologist reading mammograms5 and will soon exceed human accuracy. The patient-safety movement will increasingly advocate use of algorithms over humans — after all, algorithms need no sleep, and their vigilance is the same at 2 a.m. as at 9 a.m. Algorithms will also monitor and interpret streaming physiological data, replacing aspects of anesthesiology and critical care. The timescale for these disruptions is years, not decades.
    I will reserve my opinion of this for a few moments, but suffice it to say, it rhymes with "Wool Schmidt".

    Artificial Intelligence and Machine Learning as applied to replacing aiding radiologists were the stars of quite a number of talks and debates, and trust me, those sessions were standing room only. We old folks don't like standing through 90 minute sessions, but stand I did through several.

    One of the best was a mock debate between Drs. Eliot Siegel, who took the side of the humans, and Bradley Erickson, who insisted machine domination is relatively imminent. Of course, all physicians were supposed to be replaced in 1910 by the Vibratory Doctor...

    I can't begin to do justice to these topics, and a quick Google search will give you more information than you could possibly assimilate in a lifetime. But the "debate" made us understand that it will take the epitome of AI, Artificial General Intelligence, to begin to replace us. And THAT probably won't arrive for a long time. In fact, people who are strong believers in such things were surveyed about when they thought AGI would actually arrive. They responded:
    • By 2030:     42%
    • By 2050:     25%
    • By 2100:     20%
    • After 2100: 10%
    • NEVER:       2%
    I'm with the 2030 crowd. 

    I don't want to get into the mechanics and such of Machine Learning and image recognition and such. But some of the hype has been driven by advances in Machine Vision...Some have said that because Google can recognize a photo of a dog, it's ready to read complex medical imaging. Not quite:

    The dog is a big visual signal, if you will, but a subtle little fracture on a great big bone is only a couple of pixels out of thousands. Reading these exams is not as easy as it looks!

    Not to belabor this, but another talk from Dr. Igor Barani, founder of Enlitic, a company leveraging Deep Learning for triage, clinical support, and other non-threatening medical applications, presented some of his work, and in this video of lung nodule evaluation you can get some idea of how the machine "thinks":


    So where are we going with this? You may remember my post from last year about IBM's Watson:
    Now you might say that Computer Aided Diagnosis is already here. You would be missing the point. CAD doesn't learn. Watson, being a cognitive computer, learns. It learns the way I learned to read CT's. Hopefully it will read them better than I do. Think of it this way... I went to college to learn the chemistry and physics (and for me, engineering and computer science) needed to understand higher concepts. I went on to medical school to learn how the body is put together with all that chemistry and physiology and stuff. I learned where the pulmonary arteries were, and what happens if a clot gets lodged in one. In radiology residency, I learned how it looks on a scan if that happens. (Well, to be fair, the scanners weren't fast enough for CTPA grams back then, and so we learned the concept with conventional arteriography, but you get the idea.)

    One physician was overheard saying something like, "Bah. My first-year residents could get that one." Yes...A COMPUTER can match the achievement of a human that has gone through college and medical school. Let this sink in. Code Word: Avicenna shows us THAT A COMPUTER IN THE EARLIEST STAGES OF LEARNING HOW TO READ COMPLEX IMAGING STUDIES CAN MATCH A FIRST-YEAR RADIOLOGY RESIDENT.

    This, people, is the epitome of disruptive technology. This is a sea-change in how radiology will manifest in the future. The implications here are staggering. To me, this is MUCH more important and noteworthy than an extra Tesla on a magnet (although a Tesla in my garage would be most appreciated) or an extra hundred slices on a CT. Code Name: Avicenna represents the most important development in our field in a very, very long time. This is a fundamental change in the way we do things. It assists the radiologist, allowing him/her to perform at the highest possible level, but does not replace us. Not for the foreseeable future, anyway.

    I was right on that one, at least.

    I have seen the future, and its Code Name is Avicenna. Seriously. Trust me, I'm a doctor! But if you don't believe me, just ask Watson.
    I'll stand by every word of that. As it turns out, this was not my first article about Watson and Radiology...Back in 2011 I spoke of Dr. Siegel's efforts to train Watson. It seems our little computer has grown up.

    So where are we now?

    I spoke with several reps from IBM, and I am further reassured that HAL, I mean Watson, bears no ill-will toward us lowly humans, particularly radiologists. IBM has no plans to replace us. They said so and I tend to believe it.

    Watson himself will manifest in a few different guises, which will be deployed in the coming years. There is sort of a tentative timetable, but I was asked not to reveal that on the off chance that something comes in later than expected. Software, even intelligent software, can be cantankerous, you know. And the FDA can be even more vexing.

    You've already met Code Name: Avicenna. IBMerge today categorizes him as part of the "Watson Health Imaging Cognitive Solutions", and deems him "A cognitive physician support tool that suggests differential diagnoses options to help inform the physician’s decisions for the patient." This is the module that impressed me last year with its (OK, his) ability to call a pulmonary embolus on a CT arteriogram.  Once released to the public, well, radiologists anyway, Avicenna will concentrate on heart, breast, lung, brain, and eye problems. He will, eventually, launch from PACS as a radiologist assistant. Note I didn't say replacement. At RSNA, Avicenna was put to work in the "Eyes of Watson" display over at the Lakeside Building, chugging away at a (relatively) small palate of test cases. I didn't want to be too obvious about videoing the display, but here are a couple of screen shots showing Avicenna's on-the-fly "thinking" process:

    Avicenna has a few new peers, also named for famous old Physicians. (No, there is no Code Name: Dalai; I'm old but not famous, nor is there a Maimonides as yet.)

    Code Name: Iaso is named not for a physician per se, but for the daughter of Asclepius, the Greek g-ddess of recuperation from illness. You'll find a lot of tea-based products out there also bearing her name. She is, according to IBMerge, "(a) cognitive "peer review" tool used to detect and reconcile differences between clinical evidence and the patient’s EMR problem list and billing records with the ability to be used prospectively as well." I was told that Iaso will be looking in particular at aortic stenosis and echocardiagram results. It seems that 23% of the time, aortic stenosis is reported in the echo, but somehow doesn't make it to the EMR. Iaso will help "bridge the gaps" in information such as this.

    Code Name: Gaborone seems to be named after a town in Botswana rather than a physician (IBMerge, let me know if I'm wrong about that...) Gaborone will be "(a) cognitive data summarization tool that looks expansively at available patient data sources, filters and presents the contextually relevant information within a single view." He (I assume he...pardon my gender insensitivity) will be a stand-alone product.

    Watson for Oncology is making its mark outside of imaging. This product "(i)mproves clinical decision making by integrating disparate patient data and images in one workflow to drive evidence-based treatment recommendations." You might have seen the recent news about this Watson module saving a patient:
    University of Tokyo doctors report that the artificial intelligence diagnosed a 60-year-old woman's rare form of leukemia that had been incorrectly identified months earlier. The analytical machine took just 10 minutes to compare the patient's genetic changes with a database of 20 million cancer research papers, delivering an accurate diagnosis and leading to proper treatment that had proven elusive. Watson has also identified another rare form of leukemia in another patient, the university says.
    Not bad for a kid that never went to medical school.

    The technically-named Marktation Medical Interpretation Process may "free the radiologist to operate at the top of his/her license." From IBMerge, "Marktation is a process for interpreting medical images. When a physician labels findings on an image using text or speech recognition, the text label is simultaneously stored on the image and pushed into the clinical report. Additionally, Watson anatomical image analytics enables the text label to be posted into the right position of the clinical report and automatically adds a description of the anatomical location to the physician's label. Marktation is a reading paradigm shift aiming to improve reading speed and accuracy." In other words, this module assists us rads in marking lesions. It may sound trivial, but when you're putting little cursors on little tiny lesions and reporting them all, it gets tedious and painful. This could help. A lot.

    Finally, Watson has two other pals (siblings? cousins?) for us to play with. The Watson Clinical Integration Module "...aims to present intelligently compiled clinical information based on the indications for an exam as well as Watson's understanding of clinical relevance. This module aims at increasing reader efficiency and helping counteract some of the most common causes of errors in medical imaging, such as base rate neglect, anchoring, bias, framing bias, and premature closure."  The Lesion Segmentation and Tracking Module "...aims to automatically segment (outline and measure) physician-marked lesions, pre-mark new exams with the index lesions from prior exams, and produce tracking tables. The module aims to speed the interpretation and reporting of comparison exams in cancer patients and others patients whose findings require longitudinal tracking."

    The details of all these many faces of Watson will come with time. I predict you'll see at least some of the modules on a PACS near you sooner than you think. I could say more, but a promise is a promise.

    Nancy Koenig, General Manager of Merge (Previous CEO Justin Dearborn now runs Tribune Publishing, another Michael Ferro/Merrick Ventures acquisition, and I guess the CEO title isn't appropriate with IBM owning Merge) had this to say about our electronic friend:  "Watson cognitive computing is ideally suited to support radiologists on their journey 'beyond imaging' to practices that address the needs of patient populations, deliver improved patient outcomes, and demonstrate real-world value." And that is the antidote to the current hysteria.

    Watson, Enlitic, and all the other AI's out there, are NOT out to replace us radiologists. They are tools for us to use in our quest for ever-better patient care. Nothing more, nothing less. To fear them makes no more sense than fearing radiation, electricity, hammers, guns, or tactical nuclear weapons. Used properly, they can serve man (the last on the list works as a deterrent to other, hopefully sane folks with similar toys).

    Dr. Ezekiel and a few rather rabid AI sycophants on Aunt Minnie not withstanding, word of the demise of our profession is a bit premature. No one, and I do indeed mean NO ONE at RSNA, save perhaps for some star-struck journalists and a few companies with nothing real to show (like Deep-Something), claims we will be replaced by machines within any of our lifetimes. That is the bottom line. Watson and his cousins aren't out to get us after all.


    This situation is a wake-up call, like quite a few others we have endured or ignored over the years. Think self-referral and AMIC. AI is powerful technology, and it has great potential to help us. Could computers someday "grow self-aware and do the diagnosis themselves"? Sure, if someday has no endpoint.

    So here's my Dalai-ism on the topic, simple-minded as you might expect, but still profound, if I do say so myself:
    We need to be in control of this technology.
    What's our greatest irrational fear of AI? That it will take our jobs away. That the insurance companies or the government will latch onto Watson as a replacement for us cranky, expensive flesh-and-blood radiologists, and leave us shivering out in the cold, holding signs saying "Will Read CT For Food" and "Buddy, Can You Spare A Cup Of Barium?"

    So it occurs to me that we aren't asking the right questions. Ignore the What and When, and ask, "HOW do we keep control of this?" I posed that very question to Dr. Siegel after one of the sessions. His answer was clear: "If we are in on the development of the technology, we will have a far greater say in how it is used. And besides, can you imagine how long it will take for the FDA to approve machine reads?" And I'm sure he's right about that. And keep in mind, there are so very many other bunches of low-hanging fruit for AI to conquer. Why should radiology be at the head of the line for obsolescence? Because Dr. Emanuel hates us, apparently. Fortunately, he has no pull with Big Blue, or Deep Anything.

    So, for those fearing Big Electronic Brother, Here's my advice:  Take a deeeeeep breath, and then take a big gulp of Scotch, or a Valium, or whatever you require to climb off the ceiling. And relax. The computers are here To Serve Man.

    I'm sorry, Ezekiel. I'm afraid HAL can't do that.