BIOMEDevice San Jose 2011 – Trip Report

- by Ken McGary, KSF Labs

New international safety standards for medical equipment are on the way, namely revised IEC 60601 software validation and testing requirements, so Tuesday and Wednesday I attended the BIOMEDevice San Jose conference to get up to speed. Here’s a good background article on the new rules, which extend the concept of Basic Safety (will it directly harm me when I use it) to also include Essential Performance (for example, the accuracy of physiological reporting equipment).  Performance factors have historically been relegated to the marketplace to determine, but the ubiquity of fairly opaque and potentially fragile software systems has driven regulatory bodies to expand their scope of concern.

In addition, provision is being made for assessing the adequacy of the design process itself, in an effort to head off sloppy thinking on general principles. In essence, this makes a thorough Risk Management protocol a must, using tools like Failure Mode and Effects Analysis (FMEA), extensive user testing (a “user failure” is considered an equipment failure), and Risk Control Option Analysis. Now most of this is nothing new to serious medical device companies, but it will soon become much more rigorous, with a well-defined, intricate, and mandatory process for making sure that every last little design issue and potential “Single Point Fault” has been accounted for.

I also enjoyed presentations on human factors in medical device design, intellectual property issues, materials management, and other product development challenges. The highlight of the conference was a roundtable panel consisting of four medical device design heavy hitters: Stacey Chang of IDEO, Dr. Joseph Heanue of Triple Ring Technologies, Roger Stern of Stellartech, and Mir Imran of Modulus. Lots of war stories and sage advice was dispensed by all.

In particular, they concurred on a couple of clear trends among the medical device venture capital crowd. For one, after the 2008 market crash, investment in small initial-stage start-ups by VC firms, and thus eventually by angel investors as well, was deemed too risky and dried up considerably. The panelists expect a corresponding dearth of “ripe” biomed companies available to be acquired by the “big fish”, who in typical big fish style must continually consume the littler ones to survive. “It’s not like they are going to be developing any new technology themselves”, one of them quipped. This will conversely make those startups that do survive much more highly valued in a few years when they reach “eatin’ size”.

The second noted trend is the rise in popularity of the “one team – many companies” or incubator model, where facilities and staff are shared among several different enterprises. This leaner, more cohesive approach is now considered the standard in capital efficiency. The dot-com era standard VC pitch typically emphasized the pedigrees of the individual company officers that had been drawn to the effort. However, Mr. Imran noted, recruiting a top-quality product development team using this “star player” approach can distract start-ups for up to a year, and it can take up to another year of hard work for them to work well as a team. In this ever faster moving tech marketplace, by the time everyone gets their act together, hungry competition has likely eaten the proverbial lunch.

The sessions on clinical trials and polymer specifications were far afield from my interests, and so gave me an excuse to slip out and wander through the exhibit hall. Lots of amazing technology on display of course, much of it requiring optical magnification to fully appreciate, but here’s one of the less bleeding-edge products that I still found quite impressive…

Accelbot 500 – This ready-to-go yet bare-bones robotic XYZ stage looks to be a great starting point for automating repetitive experimental tasks. The base model will set you back $4500, but you get a roughly 500mm x 200mm x 100mm (20″ x 8″ x 4″) working volume, 0.02mm (0.0008″) resolution, a solid pre-drilled base plate for attaching your work holders, and an internal integrated controller that accepts straightforward motion commands via Python scripts or any other programming environment capable of the standard serial port protocol via USB. They were demonstrating it as a robotic pipette, but I imagined adding a small vacuum system to make a pick and place machine for loading surface mount components onto circuit boards.

No, it’s not nearly as cheap as some of the clever and awesome MakerBot kit solutions like CupCake CNC (~$1000), but it has very high precision, versatility, and ease-of-use for the price. Besides, MakerBots move the work rather than the tool head, so it’s not appropriate for many “wet” biomedical research applications, unless you don’t mind your solutions sloshed and stirred from the constant motion. I’m getting dizzy just thinking about it…

1 Response to "BIOMEDevice San Jose 2011 – Trip Report"

Leave a Comment

You must be logged in to post a comment.