Monday, September 15, 2014

Neuroinformatics 2014

Last month I attended the 2014 Neuroinformatics conference in Leiden, Netherlands. This was a new meeting for me - I'm new to the community - and it was a real eye opener. It was very helpful to be in the presence of a community who've spent so much time and energy thinking about how to manage and analyze enormous sets of biomedical data. Happily, some kind soul video taped basically the whole thing. The videos are available here:

A great set of conference photos is up on Flikr too.

Note: Albert Einstein once lectured in that very same room - and the University of Leiden has been around since 1575 I believe. Pretty amazing.

Dirichlet Process Clustering

We've been looking at Dirichlet Process Clustering recently as a great data-driven approach for learning patterns in data. This discussion is about as lucid of a discussion on the topic as I've seen.

(Note, following this link might require that you become a registered user at - which is an awesome site that you should follow anyways...)

Wednesday, July 16, 2014

Wednesday, June 25, 2014

Nicaragua - Engineering World Health Summer Institute

I recently spent a month in Nicaragua as an instructor for the Engineering World Health (EWH) Summer Institute (SI). EWH is an amazing not-for-profit whose goal is to help hospitals in the developing world have access to expertise for maintaining and repairing their equipment. EWH's research has shown that while a respectable amount of equipment is donated to hospitals in the developing world, there is practically zero support for installing, maintaining, or repairing it. As such equipment donations often fail to make the desired impact because:
  • no one knows how to install it
  • no one knows how to use it
  • the donation comes without an instruction manual
  • the instructions aren't in the local language
  • the device was donated from a 50Hz country and cannot work without modification in a 60Hz country
  • the device uses consumables (pads, hoses, etc) which aren't provided and that are hard or impossible to procure locally
  • there is no local expertise for what to do when the equipment fails
EWH seeks to fill this void in a number of ways. The Summer Institute is a program where US students travel to a developing nation for two months. During Month 1, they learn the local language and receive training on medical device operation, troubleshooting, and repair. During Month 2, the students transition to regional hospitals where they work to support the hospital's medical equipment needs in any way they can. They also perform a detailed inventory of equipment to assist EWH in forecasting how to improve their support.

This summer, EWH is operating SI's in Nicaragua, Tanzania, and Rwanda. I was the instructor for the Nicaragua program during Month 1. Now that it's Month 2, I have returned home and the students are hard at work in their respective hospitals.

Our classroom! Always hot and often noisy. A big departure from the relatively
cozy confines of Temple's College of Engineering.
This was a wonderful teaching assignment for me because it was tough and therefore I was forced to learn a lot. I have a lot of general and theoretical expertise in medical instrumentation, but I haven't personally disassembled any ventilators or defibrillators so I was fairly worried about my ability to instruct others to do so. Fortunately, I had a lot of good support, including a former instructor with a zillion years of expertise who patiently answered all my late-night email questions. I did a total of sixteen lectures, (four per week for four weeks) and we covered the following subjects:
  • Lecture 1: Basic Circuit Theory
  • Lecture 2: Electrical Safety and Transformers
  • Lecture 3: Power Supplies and Batteries
  • Lecture 4: Ventilators and Oxygen Concentrators
  • Lecture 5: Pumps, Pulse Oximetry, and Blood Pressure
  • Lecture 6: ECG
  • Lecture 7: Defibrillators
  • Lecture 8: Fetal Heart Monitoring
  • Lecture 9: Infant Incubators and Warmers
  • Lecture 10: Lighting
  • Lecture 11: Apnea Monitor and Signal Detection Theory
  • Lecture 12: Electrosurgery
  • Lecture 13: Suction and Medical Gasses
  • Lecture 14: Anesthesia
  • Lecture 15: Autoclaves
  • Lecture 16: Motors

Did I mention it was hot?

There was also a series of (very!) hands-on labs where students had to solder, wire, debug, measure, and calculate any number of circuits relevant to medical electronics. The labs started out fairly straightforward (make an extension cord) but quickly got complicated (build a power supply and use it to charge a battery). The labs were a ton of fun and a good opportunity to show the students some hands on skills.

Practicing hands-on skills during lab.

Perhaps the best part of the experience was our weekly visits to the hospital, whose full name is "Hospital Amistad JapĆ³n - Nicaragua". As the name suggests, the hospital was built through a collaboration with the Japanese government. We were told that the Japanese have done a lot of philanthropic work in Nicaragua including overhauling the drinking water supply. Good for them. But I digress. The point of the hospital visits was to give the students hands on experience taking apart, reassembling, and troubleshooting medical equipment. The engineering staff there was excellent and incredibly knowledgeable. They already know full well how every piece of equipment in the hospital works and how to repair it. We were there to learn from them, not vice versa. They were very generous with their time and expertise and soon had us doing all sorts of tasks ranging from menial to sophisticated and 100% educational.

The hospital in Granada.

Because medical equipment is expensive and hard to come by (apparently all purchases must be approved by the national health ministry...) the engineering staff takes extraordinary precautions to keep their equipment operational. Every day they take perfectly functional equipment off the floor so that it can be cleaned and serviced. Preventative maintenance is in their DNA. This ethos extended beyond autoclaves and centrifuges - they also care for their ceiling fans, air conditioning units, and refrigerators with the same devotion. 

Even the motor from a lowly floor fan can teach you a lot!

Our students took apart and cleaned a lot of air conditioners and fans. But they also got to service baby incubators, centrifuges, autoclaves, ventilators, suction pumps, and nebulizers amongst others. In each case, we'd strip the device down as far as possible and then figure out how it worked as we cleaned it and put it back together. I thought the suction pumps were especially neat since their method of operation is elegantly simple. I also learned that a nebulizer is basically as suction pump running in reverse. Sometimes the lessons came from unexpected places. By taking a fan apart, I learned what permanent split capacitor motor is, and by taking apart a motor from a cafeteria meat grinder, I learned what a centrifugal switch is and why the motor needs it (the switch disengages a startup coil once the motor spins up to speed).

Taking apart an infant incubator to see how it works. The black
element radiates heat. The white disc object on the left is a fan
that moves the warm air into the infant chamber.
Taking apart a particularly dirty suction machine. The black stuff
inside the suction chamber is dried blood! At the bottom,
you can see the rubber diaphragm attached to the motor piston.
Its pumping motion is what creates a vacuum in the suction chamber.

Watching the engineering staff fix an x-ray film developing unit! With a
little cleaning and some oil, they got this thing up and running. I took
a photography class in high school and amazingly still remembered a bit
about the film developing process, so I was able to explain it to our students
who must have thought I grew up in the stone age.

One team tried putting a broken industrial dryer back into service. One of our savvy students found an english language warning sticker on the back of the dryer indicating that the gas would automatically be cut off if the exhaust duct was clogged. Of course their engineering staff couldn't read this warning and so had not thought to check the duct. Our students took it apart and sure enough it was very much clogged! They were able to clear a lot of the clog out - the dryer still isn't working but I'm pretty sure its a lot closer to functional than it was a month ago.

Mid-morning mango break!

We had one especially fulfilling experience where we were able to repair an infant incubator which had been on the fritz for over a year. The engineering staff had already repaired it but had somehow missed one connection (which to their credit was very hard to spot). I showed the students how to draw a circuit diagram by studying the circuit board, and together we determined that one of the connections on the circuit diagram had been overlooked during the original repair. A single soldered wire allowed us to turn the machine on! Then we noticed that the fuse kept blowing. Que tal? Our on-the-ground-coordinator (and lab instructor, and former EWH-SI student) remembered that its not uncommon for people to put the wrong fuse in a device. Sure enough, the incubator needed a 5 amp fuse but someone had put a 1 amp in instead. We swapped up to a properly rated fuse and lo and behold, the thing was functional. As far as I know that machine is back on the floor now, warming babies. That was a highlight of the trip, for sure.

Fixing the faulty circuit board in the infant incubator. As you
can see, the staff had already made a repair (after the board
overheated and partially burned!). We were fortunate enough
to spot a single unconnected trace and correct it.
Analyzing the circuit board.

My only regret is that I'm not around for Month 2 to visit students and keep learning about all these superbly interesting medical devices! And I'm also a bit jealous that they get to keep practicing Spanish - I'm worried I'm going to forget a lot of what I learned :-/

One final thought ... for the first time in a long time, I got to be a student again. We had Spanish lessons four days a week for four hours. I forgot that its sometimes hard work to be a student. Its hard work to sit and listen to someone speak at you. The classes were a lot easier to digest and a lot more fun when the learning was interactive. I noticed that after about two weeks of Spanish, I would have rather stopped learning new Spanish and instead focussed on becoming proficient at what I'd already learned. I suspect there's a lesson for me there in terms of how I approach teaching engineering. I try to be an upbeat and entertaining lecturer, but there's really no substitute for getting students to practice and become proficient, even if it means potentially teaching less material.

My Spanish notebook!

So that was that. A fulfilling and exciting chapter of my professional life. I hope EWH will have me back to teach again one day. Africa? Middle East? Sign me up!

Thursday, January 23, 2014

Concussion in Sport

I found this quote sobering (from the 2012 Consensus Statement on Concussion in Sport):
There is no good clinical evidence that currently available protective equipment will prevent concussion ... Biomechanical studies have shown a reduction in impact forces to the brain with the use of head gear and helmets, but these findings have not been translated to show a reduction in concussion incidence.
Apparently helmets are handy for helping you not to crack your skull open, but not for preventing concussion. Thats a happy thought to mull over as we head towards Superbowl Sunday :)

Monday, January 13, 2014


So I recently completed my first ever MOOC, a course on Machine Learning offered by Coursera, which is a startup launched by Andrew Ng at Stanford University. I was excited to take this course because (a) I really want to know a lot more about machine learning and (b) I want to see how a MOOC works and to get a sense for its strengths and weaknesses. Are MOOCs really going to be a disruptive technology? Are they going to render brick-and-mortar universities redundant relics? Are MOOCs going to do to universities what craigslist did to newspapers?!

Here are my thoughts, in no particular order:

  1. It was a very good course. I learned a lot of useful information in a format that was much easier to digest than if I'd just read a textbook. 
  2. There were ten weeks and each week had a series of lecture videos ranging in length from 5-10 minutes. There were in-lecture quizzes (which didn't count for anything) as well as comprehension quizzes to be taken online after the quizzes. Finally there were homework assignments that required writing code in Matlab or Octave. Solutions were uploaded and graded automatically through a nice seamless interface.
  3. I  loved that the lectures could be watched online at variable speeds. I also loved that I could download the videos and watch them on my phone during my subway commute to and from work. Being able to pause, skip, and review video is a huge advantage. If you zone out for a moment and miss some important point, you can just re-watch it instead of being confused for the remainder of the lecture. Win. I also loved that I was able to download all the slides.
  4. The quizzes were often pretty challenging. Mostly they were multiple choice but sometimes I had to do numerical computations.
  5. The homework assignments were often interesting but in order to make them tractable for large numbers of students of many different backgrounds and strengths, the instructors provided an awful lot of pre-written code. This made many of the assignments feel a little plug-and-chug, even though they were tackling pretty complex real-world issues (e.g. spam detection).
I think that last issue is the real sticking point for MOOCs. For a class like this one to really be excellent, students must be challenged with big open-ended design problems that can be solved with many approaches. The student has to figure out for example, what are the best feature vectors, how to code them, how to tune the parameters, etc. Those design problems must be tough and force the student to really struggle. That in turn requires instructors or teaching assistants that can work hand in hand with the students, answering questions, making suggestions, and so on. While the course had an online message board that was supposed to play this role to some extent, I don't think its the same has having face time with an experienced instructor.

None of this is said to knock Coursera or Dr. Ng. Quite the contrary, I feel like I got a great bargain on  my investment of zero dollars (although I put a good deal of time into it). However I'm just thinking about the bigger picture of how a MOOC would have to look to be truly a replacement for brick and mortar universities.

I think another big problem MOOCs will face is that a whole lot of students need advising - what courses to take in what sequence, how to study, how to do problem sets, etc. While many individuals can sort this for themselves, a lot of folks can't, and thats a big part of the value that a university provides. Right now I think MOOCs are probably a great resource for people who already know how to study or already have some higher ed under their belts. This article from Inside Higher Ed has some interesting demographic information on who takes MOOCs.

Anyways, it was a fun course and as far as the instructor is concerned, I am now an "expert" (his words!) on Machine Learning!

Friday, December 13, 2013

Printing Prosthetic Limbs

Pure awesomeness. Technology is such a force for good in this world.