Attending AWS re: Invent, Amazon Web Services’ First Global Customer and Partner Conference: What’s The Healthcare Angle?

[CW: I promised to update this blog post after the conference, but wrote a complete new post instead: 2012 Amazon Web Services (Health) User Conference Trip Report: Sounds Like (Nine)Teen(99) Spirit]

I’ll be/I am/I was (depending on when you read this) attending AWS re: Invent, Amazon Web Services’ first Global Customer and Partner Conference on Tuesday, Wednesday, and Thursday of this week. I’ll be looking for the healthcare angle to Amazon’s vision of cloud computing.


The conference runs (ran) from 11/27-11/29 in Las Vegas. I’ll tweet about it, so I embedded a couple tweets streams below. I’ll favorite tweets for later. After the conference I’ll replace the live tweets with a retrospective of my favorite favorites. I’ll use the short URL during the conference, if you want to look up a session (deep linking to specific sessions doesn’t work, so see below). If you’re not on Twitter, stay tuned to this page. If you are on Twitter, you can follow me at @EHRworkflow.

Here are my recent tweets. During the conference they’ll likely be about the conference. Before and after (before I edit this post) you’ll see non-conference-related tweets.

Tired of listening to me? Here are recent tweets containing the #AWSreInvent hashtag. Most of these will be posted from other folks tweeting about the conference, but you may also see a couple of mine fly by. Sorry about that.

Here on some abstracts from some of the 150+ sessions that caught my eye before the conference. While AWS re: Invent is not an industry vertical-oriented health IT conference such as HIMSS and AMIA, most of the following topics have implications for computing in healthcare: cloud, workflow (or course), security, mobile, games, data science/big data, etc. I won’t necessary go to every one (Look! Something shiny!). But I’ll probably go to a lot of them. I’ll update this blog post after the conference. So check back.

Scaling Your Application’s Work in the Cloud with Simple Workflow

Dealing with scale and concurrency in today’s web and mobile services can require complex business logic in your application. To achieve high scale in the cloud, often developers have to coordinate and track state for steps in application processes distributed across remote data centers. Come to this session to learn how Amazon Simple Workflow (SWF) manages and coordinates your application sequences in “workflows” by our AWS pay-as-you-go service. We will walk through real-world examples of customers who are basing their high-scale, fault-tolerant applications on Simple Workflow today.

Security In the AWS Cloud

There are so many different thoughts about how to secure your applications running in AWS that it can be confusing to know where to start. In this session, we cover tips, tricks, and emerging best practices for securing your applications. We discuss topics ranging from how to configure your AWS resources to options for logging and intrusion detection. Discover that running your applications in AWS gives you a great head start.

Building a Mobile Application Platform on AWS

Get under the hood with’s founder to see how they used AWS to build their mobile Platform as a Service. In this session, you learn how Parse is using a variety of AWS services including Amazon EC2, S3, ELB, EBS and Route53 to build data storage, push, and easy upload services for mobile developers.

Distributing Work in the Cloud with Amazon’s Simple Workflow (SWF) and the Flow Framework

Applications today can span on-site and off-site environments, as well as across multiple compute resources in the cloud. Come learn how to simplify your application’s state management, asynchronous tasks and work distribution with Amazon Simple Workflow (SWF). During this session, you will learn how to use the SWF Flow Framework to define your application logic in “workflows” that are managed at high-scale and with fault-tolerance by Amazon SWF.

Meteor Entertainment’s Lessons Learned and Best Practices for Game Developers

Game developers need to spend their time building new games and features, not managing infrastructure. Meteor Entertainment has learned how-to minimize the time they spend managing infrastructure by automating deployments, monitoring systems through log analysis, and by making their data tier easy to scale. Attend this session to hear all about Meteor’s best-practices.

How to Deploy Python Applications on Elastic Beanstalk

(Hey! I like Python.)

Learn how to configure, deploy and scale a Python application running on Amazon Elastic Beanstalk. This talk uses two samples, a simple url shortening API built using Flask, and an image processing app built using Django, to demonstrate how to quickly get up and running on Amazon Elastic Beanstalk. In addition to learning best practices, the talk covers performance tweaks, and options for scalable data storage including S3, DynamoDB and RDS.

Data Science with Elastic MapReduce

In this talk, we dive into the Netflix Data Science & Engineering architecture. Not just the what, but also the why. Some key topics include the big data technologies we leverage (Cassandra, Hadoop, Pig + Python, and Hive), our use of Amazon S3 as our central data hub, our use of multiple persistent Amazon Elastic MapReduce (EMR) clusters, how we leverage the elasticity of AWS, our data science as a service approach, how we make our hybrid AWS / data center setup work well, and more.

Solving Big Problems with Big Data

The problem of big data is not only that it is capacious, but that it is also heterogeneous, dirty, and growing even faster than the improvement in disk capacity. One challenge is then to derive value by answering ad hoc questions in a timely fashion that justifies the preservation of big data. A group of us from databases, machine learning, networking, and systems just started a new lab at University of California, Berkeley, to tackle this challenge. The AMPLab is working at the intersection of three trends: statistical machine learning (Algorithms), cloud computing (Machines), and crowdsourcing (People). One of the driving applications for the AMP Lab is cancer genomics. Over the next several years, gene-sequencing technologies will begin to make their way into medicine, offering the most complex tests available. This advance brings a new type of data with tremendous promise to help elucidate physiological and pathological functions within the body, as well as to make more informed decisions about patient care. The cost of genome sequencing is projected to fall within range where it may be used for diagnostic and treatment purposes within the next two years. Due to the overwhelming amount of information returned by these tests, direct human interpretation is not feasible, and therefore will have to be guided by computational methods and visualization. The use of sequencing information has debuted in cancer. A provocative hypothesis is that the massive growth of online digital descriptions of tumor cell genomes will enable computer scientists to help make breakthroughs in cancer treatment, perhaps even within the next few years. Learn about the frightening fractions of cancer, dramatic speedups in genomic data processing by using cloud computing, and the blurring between opportunity and obligation when dealing with a problem that affects the lives of millions of people.

Big Data Masterclass

Learn how engineers at startups and larger enterprises use data to drive greater insight into their operations, customers, and business in this lively discussion of big data techniques and tools. From Hadoop to data warehouses, this panel discusses the tools, techniques, tips, and tricks for building data driven teams and delivering cost optimization at scale.


A Great Workshop about Diagnostic Error: A Multidisciplinary Exploration

[The following tweeted, about an upcoming workshop in Long Beach, was added 2/3/2013.]

I had the very good fortune yesterday to attend a great workshop on Diagnostic Error: A Multidisciplinary Exploration.

The workshop was held at the Health Sciences and Human Services Library (@HSHSL), at the University of Maryland Baltimore. It’s an impressive facility. But I love history. Across the street is Davidge Hall (@ummedschool), the oldest building in the country used for medical education.


Diagnostic Error: A Multidisciplinary Exploration was taught by the highly knowledgeable communicators:

From the prospectus:

“Diagnostic error, a major factor in patient harm also increases medical costs ( A free, interactive workshop for medical decision makers and information professionals is being held on November 15th focusing on how multidisciplinary teams can contribute to the reduction of diagnostic error. This innovative session will highlight evidence-based processes and the collaborative roles of clinicians and their librarians/ informationists as they work together to reduce factors contributing to diagnostic error.”

The session covered:

  • “Team-oriented approaches to understanding the role of information and evidence in the diagnostic process.
  • Case analysis and discussion of bias.
  • Partnering of librarians/informationists with clinical staff to strategize improvements
  • Application of failure analysis techniques to explore system and process improvement.
  • Design of evidence sharing innovations to reduce diagnostic error.
  • Strategies for implementation of proposed projects.
  • Multidisciplinary teams from organizations are encouraged to attend. MLA CE credits will be provided.”

While I did not live-tweet the workshop per se, I did find and tweet some excellent, freely available on-line resources mentioned in their slides. The purpose of this post is to thank Elaine, Mark, Barb, Linda, and Lorri; to archive those tweets and related material (see below); and to encourage others to attend future instances of this course.

Here’s a bit more about Davidge Hall.


Thank you all! I learned a lot. I understand this was the second time you taught this course. It felt like you had vetted it many more times than that. I hope anyone who reads this or hears of an opportunity to attend Diagnostic Error: A Multidisciplinary Exploration will do so. You won’t regret it.

And if you simply can’t make it, keep an eye for the #dxerror hashtag on Twitter. You’ll probably find one or more of us hanging out there!

And thank you Susan Carr, Editor of Patient Safety & Quality Healthcare, for the following hat tip on Twitter.

#AMIA2012 Trip Report: Best Conference I (N)ever Attended

[UPDATE: @AMIAinformatics kindly mentioned this blog post in two newsletters! 11.15.12 and 11.21.12. Thank you @AMIAinformatics. Also, since this post I’ve changed my Twitter name from @EHRworkflow to @wareFLO.]

I’ve live-tweeted lots of conferences, but I never live-tweeted a conference I didn’t attend. Until #AMIA2012. #AMIA2012 was the Twitter hashtag used during the recent American Medical Informatics Association Conference in Chicago.

When AMIA published their program, I published a blog post (EHR Usability, Workflow & NLP at AMIA 2012: Presentations You Don’t Want to Miss!) listing the tutorials, workshops, panels, presentations, and posters that I wish I could attend. Why? To get people to attend them and tweet so I could listen from afar.

That’s all I thought I’d do: listen. Ha. Check out the #AMIA2102 social analytics…


I was one top five #AMIA2012 twepes for mentions, tweets, and impressions (tweets times followers). That’s me, @EHRworkflow, in the “HatCam”.

Twitter has a neat feature of being able to embed tweets into a blog. I embedded the most interesting tweets (from my point of view) into my blog and wrote this post around them. For another example of this, see my Clinical NLP at 2012 NAACL Human Language Technology Conference.

“Two informaticians walk into a bar…” and This Blog Post Walked Out

The above tweet was a spark in a dry forest. Whoosh! Out of shear megalomaniacal grandiosity, I’ll include some of my own best efforts below. On November 7th, in Chicago, Illinois, before an assembled horde of medical informaticians, was unveiled the winner of the “Two informaticians walk into a bar…” Trademark Infringement Award:

Many thanks to @matlatzinca for capturing this moment for posterity. Yes, this was my personal favorite too! As well it was of many other #AMIA2012 attendees (see below). No, I am not bitter that another joke won the grand prize. To each their own. Humor is so ephemeral, so idiosyncratic, so, so, well, undrinkable (dratted spellchecker! Unrankable! Even spellcheckers have opinions these days!). But let me list some of the accolades for this joke, my joke, from some of the brightest lights in medical informatics today.

Here’s a few more of my other little gems. I promised @AMIAinformatics not to steal their thunder by publishing more than my own jokes. Fine with me. If you don’t get a joke, feel free to contact me.

Sorry @AMIAinformatics and @ekrub! Needed to set up my next (admittedly feeble <- BZZZZT! FALSE MODESTY ALERT!) joke.

Heh! See who I’m hanging with? “Love”!

I had my critics…

…and my defenders.

For all xs and ys, where x is an Informatician and y is an Informatician and x and y aren’t the same…oh, just forget it!

Oops! I need to move that closing outer bracket all the way to the end to scope the x and y variables past the walkIn2 predicates. This one won the Most Technically Appropriate (Super Geek) Award!

And this one won the Best Collaboration Award, with @techydoc.

Believe it or not, these stellar gems were just a tiny percentage of the extraordinary firmament that was “Two informaticians walk into a bar…” When the rest are showcased by @AMIAinformatics, I’ll add the link here.

Workflow and Usability: Pssst! Can I Interest You In A Blog Post?

“Workflow” and “usability” were big topics at #AMIA2012. (Check out the wordle, center-right) I knew they would be. Why not? I see so many headlines about EMR and EHR workflow and usability problems. Are they impeding adoption? What can be done? It’s a target-rich environment for medical informaticians interested in these subjects.

Which brings me to the question: Why bother?

Why did I spend so much time and energy following, tweeting, and writing this? I do have an agenda and I’ve found Twitter to be an excellent means to execute it. This blog has over a hundred posts, white papers, and presentations about EHR workflow and usability. I collect (“curate” in recent terminology) links on ehr workflow, usability and related subjects and publish them to a single-page, Drudge-style EHR.BZ Report on EMR / EHR Workflow, Usability, Productivity and Safety. Over 1600 links so far.

While following and tweeting and retweeting, I look for opportunities to slip a few of my links into the tweet stream. The jokes? A spoonful of sugar, so to speak, to help the medical informatics go down. I won’t belabor my views here, though I do hope you wander off into my blog to be belabored of your own accord, and therefore your own blame too. But I will include some of those “Pssst!… Can I interest you in a blog post” tweets here. By the way, lots of my tweets were retweeted (RT’d). The RTs don’t show up below unless they were old-fashioned RTs (manually created, sometimes to include a prepended comment).

So, my tweets are a bit like this blog post: subjects of general interest, to an admittedly specialist community, wrapped around serious stuff I aim get out to influence thinking about EHR workflow. (Isn’t there some metaphor involving Oreo cookies or pills in hamburger that applies here?) Anyway, back to a subject of increasing general interest: clinical natural language processing!

Clinical Natural Language Processing: Yup! … Another Blog Post.

In this case “Another Blog Post” refers to what you can do with tweets after they are live-tweeted during an presentation. They make a great outline on which to hang a subsequent blog post. This “Trip Report” blog post about attending AMIA 2012 via #AMIA2012 is *not* a good example. Wasn’t there; bounce around too much; not my purpose.

However, take a look at the preamble from blog post based on a more disciplined series of tweets from another recent conference.

“My goal was threefold:

  • Leverage social media content I went to some effort to create (the tweets).
  • Summarize current state-of-the-art clinical NLP research and directions.
  • Make it understandable to readers who are not computational linguists or NLP engineers.”

So you can imagine how pleased I was to find, amidst the 4500+ #AMIA2012 tweets and 500 or so #AMIA2012 tweeters, an absolutely exemplary series of tweets devoted to the Introduction to Clinical Natural Language Tutorial at #AMIA2012.

What I really, really like about the following sequence of tweets from@Sorena997 is that they meet all the criteria for live-tweeting a conference presentation. They have clear beginning and end. The beginning clearly establishes context. The end clearly summarizes the most important take-home point. In between, the tweets are easy to read, with correct grammar within Twitter’s space constraints. The tweets are uninterrupted by extraneous RTs and comments. They provide important keywords one would use to search for more information. While watching the tweets I had a second browser open in which I was Googling all sort of stuff: Polysemy, Negation, Uncertainty, Temporality, Implication, Wendy Chapman, Leonard D’Avolio, NLP pipeline, and so on. No, it wasn’t as good as being there. But I got a lot out of it. I look forward to attending such a tutorial, in person, sometime soon.

What I’d do next, if these were my tweets, is I’d write a blog post around them. Just like I did with my Clinical NLP at 2012 NAACL Human Language Technology Conference. That’s just me. Writing about areas you’re interested in is a great way to learn about those areas. (One of my favorite books is Writing to Learn.) In fact, I was tempted to so anyway. Then I decided it would take away from their elegant narrative arc — and my main point. (Which is? Hmm. Something about Twitter and blogging and medical informatics conferences go together like a horse and carriage and… That’s not going to work. Never mind.)

So, without further ado, I give you at a good example of how to live-tweet an informatics presentation.


OK! Phew! If you think I’m a bit obsessive, you’re half right. I’m obsessive about this stuff (whatever that is — intersection and interaction between medical informatics and social media?). But if you ever meet me, I’m very laid back. I look forward to proving it to you someday.

I’ll bet you didn’t think this conclusion was going to be as short as it is!

Peace, out!