Looking at bioengineering measurements courses

In  Giving up on light-based pulse sensor, I discussed the difficulty I was having in finding comparable courses to borrow ideas for labs from. In the comments kcab suggested “I would look first under Biological Engineering and in general look for instrumentation and measurement courses.”  I started doing that today.  That does seem a more productive direction for the search.

• At George Mason University, the syllabus for BENG 301: Bioengineering measurementslists two textbooks:
  • Medical Instrumentation: Application and Design, 4th Edition. John G. Webster. Publisher: Wiley, 2009. # ISBN-10: 0471676004; # ISBN-13: 978- 0471676003.
    I looked at the table of contents of the 3rd edition on Amazon, and it looks like a promising collection of material.  The 4th edition is ridiculously over-priced, but the 3rd edition is a reasonable price ($23 used, $44 new).  Several of the topics covered are ones we had already planned on (temperature and EKG, for example).  I had thought a little about measuring breath, but not about blood pressure or flow rate.  We are definitely limited to non-invasive techniques, so I need to get a copy by Inter-Library Loan to see if there is much more that we can use.  It doesn’t look like this book covers basic circuits, though it does cover op amps, so it is probably not suitable as out main text.
  • Introduction to Biomedical Engineering, 2nd Edition John Enderle, Susan Blanchard and Joseph Bronzino, Publisher: Elsevier, 2005.
    Since there is a newer edition of this book, the 2nd edition is also reasonably priced.  Only Chapters 8–11 are relevant to the course we are planning, which seems like too little of the book to justify purchase.
• The accompanying lab course BENG 302 (same instructor) has a syllabusthat lists 10 labs:
  1. Data Acquisition Basics
  2. Statistics
  3. Electrocardiography
  4. Electromyography
  5. Electroencephalography
  6. Electro-oculography
  7. Blood Pressure Laboratory
  8. Pulse Oximetry
  9. Glucose Monitoring
  10. Open laboratory – Student Design Lab

  I don’t have detailed information on the course, but I get the impression that the students are using existing equipment, not designing their own, so the labs are not really that relevant to the course we’re designing.  I should probably contact the instructor (Joseph J. Pancrazio) and find out what he is doing in the course and lab—there may be something there that we can borrow.  I think that our course will be ending where his begins (with electrocardiography and electromyography).

• CMPE 491/691 Biosensor Technology (syllabus) at University of Maryland Baltimore Campus uses John Veteliono and Aravind Reghu, Introduction to Sensors, CRC Press Taylor & Francis Group (2010).  The table of contents seems heavy on theory and light on wiring up the sensors to do things.  They do mention several interesting sensors though: chemiresistors, thermistors, and fiber-optic sensors.
• There are a lot of bioinstrumentation courses, but they are almost all about using existing instruments, not designing the instruments.
• MIT Biological Engineering II: Instrumentation and Measurement
  

  Electronics:
  Resistive Networks, Filters, and Op-amp Circuits for Measurement

  During the first part of the course we will focus on electronics. Over a series of labs, we will build several types of commonly used electronic circuits and combine them implement a system for measuring DNA melting curves. This section will also provide an introduction to computer control and data acquisition, including LabVIEW and MATLAB® software.

  Measuring DNA melting curves sounds like a good follow-on to the thermistor lab.  In the labs section of the web site they mention using photodiodes to measure fluorescence to detect the DNA melting. As with many of the Open Courseware offerings, the important details are not included in the web site. But Open Wetware explains the technique (and that seems to be the actual lab from 20.309 course). There is also a packet of most (all?) of the handouts for the course.  There’s a newer copy of the lab manual on the MIT web site, not linked to by OCW.
  It looks like an excellent lab for our course in terms of content, but we would need to build the box and optics for the students, leaving just the LED, photodiode, and amplifier for the students to build. The photodiode amplifier may be messy to build (much smaller signals than the EKG, so even more noise problems) and we don’t really have a lab with an optical table, wet lab tools for measuring out the DNA, and oscilloscopes. We could probably have the students measure the DNA into cuvettes in one lab and carry it to another lab for the electronics, but I don’t know that we could get the lab tested and enough copies of the block and optics built and tested in time to run the course this year.  This definitely needs more thought!

My search also found another idea for a lab—force and contact area measurement, which are used in studying stance and gait, for example.  The TekScan FlexiForce sensors are a bit pricey ($15 each in quantities of 8) though the conductance output (linear with force) is an interesting twist on a resistance output.  TekScan’s arrays of pressure sensors look like a cool technology, but there does not seem to be any prices associated with them, which means that either we have to contact the company or not use them. They also would require multiplexing to scan the arrays of sensor, which may require more digital design than we want for this course.

In any case,  I prefer teaching students how to use commodity parts, rather than custom components that may not exist in 6 months or a year.  Commodity force-sensing resistors are cheap ($7  from Adafruit, $8 from Sparkfun, $5.30 from Digikey) and have an interesting power-law relationship for resistance (or conductance) to force.  It may be worth incorporating one of them into a lab, though they are not very accurate as measuring instruments (10% accuracy).

Note: I thought for a bit thatcheap alcohol sensors were chemiresistors, but they are not—they’re fuel-cell sensors, producing a current output. Forensic Issues in Alcohol Testing by Steven B. Karch briefly mentions fuel-cell sensors, but says that forensic sensors generally use infrared absorption at 3.4 or 9.5μm, as infrared light sensors are easier to calibrate and test, making them more useful in court.