Code Name: Avicenna The Future In Progress

One of my Radiology professors back in residency, a very wise man, had a saying: “The more dogmatic you get, the more likely you will be wrong.”

In the medical business, there is a tie for the three most important little words: “I was wrong,” competes nicely with “I don’t know.” (If you were wondering, the four scariest words in the radiological lexicon are: “You read a scan…”)

The Future has a way of sneaking up on us, and occasionally biting us on the behind. I always thought, for example, that age 50 was a long way off. Now I’m well into that decade of life, and the 60’s are looming. As Steve Miller put it, “Time keeps on slippin’, slippin’, slipping’, into the Future…”

All this leads us to the fact that I was dead wrong about something futuristic, something I thought we wouldn’t see until many years from now. Something I saw this week at RSNA at the Merge booth. It is a Work In Progress titled “Code Name: Avicenna”, a peek into the future at some very disruptive technology (I use that term with all due respect and awe) brought about by the new consortium between Merge, now a wholly-owned subsidiary of IBM, and IBM Watson Health.

Let us speak a moment about Avicenna, whom I’m assuming is the inspiration behind the Code Name: Avicenna project. Since you are reading my blog, you are of course quite intelligent and well-educated, and thus you have probably heard of Avicenna. I, however, had not, so I turned to the Wiki:

Avicenna (c. 980 – June 1037) was a Persian polymath who is regarded as one of the most significant thinkers and writers of the Islamic Golden Age. Of the 450 works he is known to have written, around 240 have survived, including 150 on philosophy and 40 on medicine.

His most famous works are The Book of Healing – a philosophical and scientific encyclopedia, and The Canon of Medicine – a medical encyclopedia, which became a standard medical text at many medieval universities and remained in use as late as 1650.

Besides philosophy and medicine, Avicenna’s corpus includes writings on astronomy, alchemy, geography and geology, psychology, Islamic theology, logic, mathematics, physics and poetry.

Pretty amazing guy. Now personally, I would have gone with “Code Name: Maimonides“,  after Moses Maimonides, an equally famous physician of the middle ages, or at least with his acronymed nickname Rambam (for “Rabbeinu Moshe Ben Maimon”). But then this isn’t my project, is it?

The choice of the code name for this Watson-based process clearly tells us where we are going; Watson is learning medicine, and doing so at a very young age, as did Avicenna who became a physician in his teens.

IBM Watson, image courtesy IBM.com

Everyone has heard of IBM’s Watson. Watson thinks, or at least simulates it nicely:

We produce over 25 quintillion bytes of data everyday and 80% of it is unstructured. Therefore, it’s invisible to current technology. IBM Watson is a cognitive system that can understand that data, learn from it and reason through it. That’s how industries as diverse as healthcare, retail, banking and travel are using Watson to reshape their industries. Watson is designed to take data in all its forms—including unstructured—and understand it, reason through it and learn from it. In a sense, Watson can think. When Watson thinks with you, you can outthink.

I cannot proceed without mentioning the old joke about the movie 2001: A Space Odyssey. The rogue self-aware computer, “HAL 9000” was supposedly a joke on IBM, the acronym being one letter off, yes? Arthur C. Clarke (or was it Stanley Kubrick?) denied this, claiming that it stood for Heuristic ALgorithm, and in fact IBM helped considerably with the movie details.
And this brings us to IBM’s acquisition of Merge, which I blogged about when it was announced. It has since been finalized:

IBM itself whet our appetite for what was and is to come on the Watson Health website:

IBM today announced that Watson will gain the ability to “see” by bringing together Watson’s advanced image analytics and cognitive capabilities with data and images obtained from Merge Healthcare Incorporated’s medical imaging management platform. IBM plans to acquire Merge, a leading provider of medical image handling and processing, interoperability and clinical systems designed to advance healthcare quality and efficiency, in an effort to unlock the value of medical images to help physicians make better patient care decisions.

Merge’s technology platforms are used at more than 7,500 U.S. healthcare sites, as well as most of the world’s leading clinical research institutes and pharmaceutical firms to manage a growing body of medical images. The vision is that these organizations could use the Watson Health Cloud to surface new insights from a consolidated, patient-centric view of current and historical images, electronic health records, data from wearable devices and other related medical data, in a HIPAA-enabled environment.

I can’t quite shake the thought of HAL, I mean Watson, watching us, but in a good way:

While it is just a Work In Progress, Code Name: Avicenna presents the first steps in realizing the unified goals of teaching Watson to see radiologic images (among others) and putting that, um, knowledge to good use for our patients. And you can see it in action today. Here is Merge’s description of the demonstration, i.e., Code Name: Avicenna…

Merge PACS™ workstation viewer and IBM Watson Health – a vision for how to help radiologists with clinical decision making

Radiologists and cardiologists today have to view large amounts of imaging data relatively quickly leading to eye fatigue. Further, they may have limited access to clinical information relying mostly on their visual interpretation of imaging studies for their diagnostic decisions.  In this demo, we present a futuristic workstation for radiologists where their normal viewing of imaging studies is augmented with clinical and imaging summaries to help their clinical decision-making. This technology could assist by running in the background to collect relevant clinical, textual and imaging patient data from electronic health records systems. It could then analyze multimodal content to detect anomalies and summarize the patient record, collecting relevant information pertinent to a chief complaint. The results of anomaly detection would then be fed into a reasoning engine which uses evidence from both patient-independent clinical knowledge and large-scale patient-driven similar patient statistics to arrive at potential differential diagnosis to help radiologists’ clinical decision making. Compact clinical summaries, along with the findings from imaging studies, would be available both for simultaneous viewing and export as a DICOM SR report.

The demo will show our vision of this futuristic technology using the Merge PACS™ workstation. The radiologist will open an exam from the Universal Worklist (UWL).  When the exam is opened, both the PACS viewer and the IBM Watson Health work in progress will be launched in separate monitors to show respective content. The ultimate output from the tool in the form of a pre-populated radiology report will then be presented to the radiologist to review and consider in making his or her decisions.

DISCLAIMER: The capability demonstrated here is for DEMONSTRATION PURPOSES ONLY. The capability is in the research and development phase and is not available for any use, commercial or non-commercial. Any statements and claims related to the capability are aspirational only. The case study in this demonstration is a hypothetical case study using fictitious medical information and do not represent an actual medical case. The results contained in this demonstration were obtained in a controlled environment and represent a vision of possible future technology. The demo will show our vision of this futuristic technology using the Merge PACS™ workstation.

The punch-line to Code Name: Avicenna is quite simple. IT WORKS. A case was presented to Watson consisting of history, physical findings, lab values, and a CT. Well, it was a CT-Pulmonary arteriogram, so Watson had a little clue there. The demonstration progressed to show Watson’s integration of the data into a cloud display of likely diagnoses. He, OK, it, proceeded to analyze the CT, showing outlines of his its regions of interest. And Watson found the majority of the emboli on the very positive scan. His ROI’s matched those of the training radiologist quite well. And it then displayed dozens of priors from its memory with similar findings. The most likely differential correctly became pulmonary emboli, which was of course correct.

As an aside, many have wondered just how Watson acquires the images upon which he it trained. I had asked this question of the Merge execs early on, but they weren’t ready to answer, until now. Basically, the scans are collected with secondary use rights, to which the institutions providing them must agree. The images and reports and other data are anonymized, so there is no privacy problem. To date, several big name operations have signed on to this effort, including Johns Hopkins and others you might have heard of. I’ll be glad to sign mine over, too. There simply is no downside to doing so. There are 30 BILLION images in Merge’s iConnect cloud service already. That should keep Watson busy for a millisecond or two.

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 much 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 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. 

via Blogger http://ift.tt/1SASANs December 03, 2015 at 04:00PM

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s