In the survey of more than 300 hospital-based physicians, only one quarter of the surveyed physicians report access to a spirometer; in addition, physicians also highlighted their lack of training in use and interpretation of spirometry.

While the authors suggest that “spirometry is a cost-effective diagnostic tool for evaluation of lung function and for case-finding in a resource-limited setting” they offer no references on that front, and the article has no economic analysis of its own. The report sticks to the topic of the survey throughout. Hopefully some empirical data on the costs and benefits of spirometry in these settings isn’t far behind.

The article by Desalu and colleagues is entitled, Evaluation of current knowledge, awareness and practice of spirometry among hospital-based Nigerian Doctors. A provisional PDF is available here.

As noted in the previous post, we have been having difficulty cutting the polycarbonate capillaries due to their tendency to fuse together from the heat generated during the cutting process. This week, we tried cutting the tubes to an appropriate length using various new methods, and here are some pictures showing the results:

The results of cuting our capillaries using various tecniques

In our first attempt, we switched to a coarser saw which allowed us to use water coolant, though it resulted in a very rough cut. We tried to smooth the edges with heating, however this was unsuccessful (1). Note the side view demonstrating the burrs and the roughness from using the coarse blade (2). Finally, we filled the capillaries with water and froze them before cutting with the coarser band saw. The ice supported the tubing and burring was significantly reduced, though considerable debris remained from the cut (3).

We are currently working on clearing the capillaries of debris as well as keeping a look out for alternative cutting strategies. Once we have produced the capillary system, we will pursue testing its ability to produce laminar flow in our spirometer.

We began searching for materials that met this requirement, and realized that standard hematocrit tubes have appropriate dimensions. We ordered a pack of the polycarbonate style (hematocrit tubes are also generally available in glass) and tried updating our design to incorporate these tubes.

Hematocrit capillaries in place

Unfortunately, polycarbonate is a very poor heat conductor, and our attempts to cut the tubes with a fine-blade band-saw resulted in fusing our capillaries shut. As is, we are still looking for suggestions of capillary materials or any premade capillary systems that we could simply cut and plug in to our spirometer.

Ends of capillaries fused by cutter

On the software end, a team of interns has been actively refining computer algorithms that take an array of flow data points and calculate the various spirometry test values. The team is using the raw curve data from the NHANES III and comparing their calculated values with those calculated by NHANES. This process has been challenging as well, as NHANES has used data processing techniques that are not clearly defined to calculate the lung function indices. Our team is currently working on developing a program that will run numerous types of processing techniques, ranging from very low-level to advanced, on the data and compare which methods yield the results closest to those printed by NHANES.

The final front of development is focused on the internal circuitry components of the spirometer. Currently, we are working with the PIC18F13K50 microcontroller which will be responsible for taking the data from our iLite signal conditioner and passing it to the computer via USB. Because USB is an advanced and highly-standardized communication protocol, we’ve been devoting considerable time into learning how to program the microcontroller to successfully identify itself to the computer and then proceed to stream data to it. Also, although our circuitry will have relatively few components, all of them have specific electrical specifications that need to be met by incorporating appropriate accessory components (resistors, capacitors, etc.) into our design. Progress in this area will hopefully allow us to create a Printed Circuit Board (PCB) of our circuitry design by the end of summer.

The multi-faceted nature of the spirometer project has allowed the team to learn a great deal in a wide variety of fields, but we are still far from experts in any one area. Feedback on our design, including critiques of existing work and suggestions for future progress, are always appreciated!

Peter and Bug Labs have been a continual source of inspiration to me and I’m very much looking forward to watching this new project develop. He’s invited feedback on his blog so please take time to visit and contribute.

We would like to send this document to a software developer who could help us begin to code it. This document is only the first draft, and any input would be appreciated and would result in better software being produced.  Please comment if you have any suggestions, or email a revised document to openspirometry@gmail.com

Thanks for your input!

• We revised our design from the Venturi design, and we now need to decide whether to build a Lilly or Fleisch type spirometer. Both are currently used in commercial models. We are examining the pros and cons of each design and hope to have a basic prototype of each built by the end of the week. Feedback on the design features and construction is welcome as always. Find out more about these spirometer designs here:

Fleisch – http://spirxpert.com/technical2.htm

Lilly – http://spirxpert.com/technical3.htm

• After much consideration, we have decided to pursue Adobe AIR as the platform for our software. We believe that AIR will provide the smoothest path to a highly-functional graphic interface that is capable of displaying all of the data we want. However, AIR does have its disadvantages, including a large RAM requirement and no direct access to USB devices. We are examining the use of the Merapi Project software to acquire our data through a Java application (allowing USB access), but would be interested to hear other alternatives.  Here is our rationale for choosing Adobe AIR over Microsoft Silverlight, Processing, and Java:

Design matrix for software decision

• We are currently working on correlating input voltages with the output of the ZMD 31014. This process will allow us to correlate the data we see on the screen with the differential pressure recorded by the sensor. Our current setup has been giving unexpected data, but we hope to resolve the problems soon.

Info we need

We want to make our product as functional as possible for our customers, so we want to know as much as we can about the operating environment for our spirometer. Information about the following topics would be greatly appreciated!

• What types of cleaning solutions are available or commonly used in your clinics? Ethanol, Cidex, other?
• How much RAM do your computers generally have?
  • We plan on creating a graphically intensive program to show video and animation and are concerned about the capabilities of the users’ computers.
• Do patients tidal breathe through your spirometer prior to forced exhalation or do they place the spirometer to their lips after their max inhalation?
  • If you have used both types of procedures, which do you prefer? Why?
  • We would prefer NOT to have patients tidal breathe through our spirometer to reduce the risk of cross-contamination.
  • Also, we want training videos to portray the use of our spirometer accurately and ensure that it is consistent with most commercial spirometer procedures.

Thanks for your help! Please send any comments or suggestions to  openspirometry@gmail.com, or leave a comment on this post.

Of course, please don’t forget to add to our wiki as well.