Andrew Jacobson and John Schmitz
Fall 2015
VIST 405
Virtual Reality Pediatric Pulmonary Function Tests
one thing we both had in mind, virtual reality. Regardless of the medical direction we decided to
take, we wanted to incorporate virtual reality into our creation. The idea we chose to produce
was a virtual reality interactive experience that was designed for children required to perform
pulmonary function tests. These tests are required for all ages to examine lung performance in
terms of volume and air flow. The machines used are sometimes intimidating for young children,
however, which could cause issues during trials. Thus, we aimed to ease the process for this
demographic, making it both more enjoyable for the user and reliable for the tester.
We conceived our initial idea based on previous personal experience with our medical
subject. Growing up with chronic lung disease, Andrew had to undergo pulmonary function tests
often to check the progression of my illness. Many children with asthma or other lung related
issues have to perform these tests. The tests occasionally are equipped with a small animation
that appears as if it was created in the mid 1990s in Microsoft Paint to help incentivize children
to perform them (ex. “balloons” or “candles”). However, the tests have long passed their
effectiveness and need a massive update. Andrew’s mother, who is a pediatric physician, faces
this issue regularly; children are reluctant to perform reliable respiratory tests.
! We aimed to create a more realistic, immersive environment for children to help
incentivize them to provide more reliable statistics and be more willing to do them period. In
addition, we designed our interactive system to add some gamify the procedure to act as a form
of respiratory “training” or “therapy” to help produce greater results over time for all ages of
patients.
Our initial research into the subject revealed some interesting findings. We first looked at
overview of how pulmonary function tests (PFT) are performed and operate, the challenges
involved and what happens after the tests. Aside from the information we discovered on the
process itself, we also learned that children are often not willing to put forth the effort to produce
reliable, consistent results (Seed). We found a study that was performed among roughly 100
children that found that children that used animations still had less consistent results over long
periods of time. The article we read on the study made it seem that it is the interactive game’s
fault for the results, however a separate study was performed that showed how universally the
performance often progresses past what is accountable for unknown reasons (Nystad). This
suggests that it is not the animations fault, but something completely irrelevant to the first study,
and it is only natural to have diminishing performance tests over time (Neale). The last study
involved two separate group of adults. One group did not do anything different than their
standard PFTs, while the other group used a game to help “train” for PFTs. The group that used
the game far outperformed the other group that only did PFTs; giving the notion that interactive
visualization actually promotes the improvement of lung capacity and function among patients.
There were other people that had conducted research into this idea, yet had not developed it fully
nor went the direction we are opting to take, but still showed some initial promise (Bingham).
Our objective was to create a virtual reality program using Unreal Development Kit that
was integrated with Oculus Rift and a custom-built electronic spirometer. We aimed to build two
demos this semester. Initially, we planned to have one consist of a sailboat that would be
propelled across a lake based on how the user performs. The other would simulate the big bad
wolf, blowing down a little pig’s house. We chose these two, since they were fairly gender
neutral in design. However, we changed our mind on the latter, since it proved to difficult to
accomplish within the time frame allotted. Instead, we decided to update the cake level that was
one of the original levels. Some of the other levels that were discussed as potential options
included: 1) a rocket level where the player had to perform better to get farther into space and 2)
a level where the player rode a horse and had to blow into the spirometer to leap over hurdles.
We produced prototypes for all of these concepts, except the “little pigs” level. These levels
might be further explored and or developed later.
The spirometer was created using an Arduino board and a compatible bread board, along
with the following materials: PVC pipe, plastic tubing, glue, a Honeywell ASDX Ultra-low
Silicon pressure sensor, a micro USB cable, and wires for circuitry purposes. The general design
of the electronic spirometer was based off of the design of an electronic spirometer of a PhD
biomedical engineer student at Vanderbilt University . The code we used was also a modification
of the same person’s code. In order to get the code to interact with Unreal Engine 4, we utilized a
free plugin that provided minimal, yet sufficient, communication between the two softwares.
We performed a proof of concept trial that tested the general concept. In the test, we had
the user blow into a paper bag that emulated the electronic spirometer, while wearing an Oculus
Rift that played videos timed with the action of blowing into the bag. The trials provided ample
results that helped change the direction of our project slightly, including what projects we
decided to ultimately develop.
Following the testing, we completed our cost effective spirometer and continued the
improvement of our project. We first focused on the general performance of the games, and then
proceeded to progress the graphics of each. Unfortunately, the computers we worked on
throughout the semester could only handle medium quality lighting, but generally was able to
handle the high input frequency of the Arduino. That being said, one of our biggest issues we
faced was the crashing of the program, which I believe was related to a potential memory leak in
the Arduino code. We weren’t able to fully resolve the issue, but got it to a working condition.
Other issues we confronted included the learning curve we had to overcome involving both
Arduino and Unreal Engine 4, getting the plugin to operate correctly and resolving logic related
code issues.
During our initial presentation, our programs worked effectively as predicted, minus a
few crashes that we experienced. At the VIST show, however, we faced unforeseen issues. Our
values that we had programmed to affect at what levels the candles would blow out started not
registering. It took some time to resolve the issue, which involved setting the numbers at much
lower levels. After that was fixed, however, the overall reception regarding our project was great.
Users appeared to thoroughly enjoy using our product. They seemed to be more enthused by the
candle demo over the sailboat demo, since there wasn’t an observable goal in the latter. Other
criticisms/recommendations involved there being a lack of birthday related objects in the “cake
scene” and creating markers to judge the distance traveled in the sail boat level. People were able
to infer what they were supposed to do fairly easily with little instructions given, which is
beneficial for mass installation purposes. That being said, knowing the issues with our current
configuration, the next step would be to determine where our development goes from here.
Since Andrew plans on continuing the research and development of this project, he
believes in modifying the demos we currently posses into more finalized formats. Following the
completion of those two, he will perhaps create another demo for testing purposes, maybe even
one of the demos we chose not to develop listed above. There is also the discussion of going to
Startup Aggieland in order to pursue a business related venture. Hopefully, in doing so, we would
be able to be placed in contact with individuals who have more expertise in the medical or
biomedical engineering field. Regardless, the development of our project went sufficiently well
enough, considering the amount of technical issues we faced throughout the process. We believe
that our concept has a lot of potential and can be carried to a more finalized development point,
which we hope is possible to achieve in the coming months or years.
Work Cited
Bingham, Peter M., Thomas Lahiri, and Taka Ashikaga. "Pilot Trial Of Spirometer Games For
Airway Clearance Practice In Cystic Fibrosis." Respiratory Care57.8 (2012): 1278-1284
7p. CINAHL Complete. Web. 16 Dec. 2015.
Neale, A V, and R Y Demers. "Significance Of The Inability To Reproduce Pulmonary
Function Test Results." Journal Of Occupational Medicine.: Official Publication Of
The Industrial Medical Association 36.6 (1994): 660-666. MEDLINE Complete. Web.
16 Dec. 2015.
Nystad, W., et al. "Feasibility Of Measuring Lung Function In Preschool Children."Thorax 57.12
(2002): 1021-1027. Academic Search Alumni Edition. Web. 16 Dec. 2015.
Seed, M.Sc., RCPT(p), Laura, Allan Coates, M.D., Susan Carpenter, RN, and Jennifer Leaist,
RN, BScN. "Pulmonary Function Testing." Aboutkidshealth. Sick Kids, 17 Mar. 2010.
Web.
