Earlier this month, we had the opportunity to visit several hospitals to collect feedback on the PneumaShoe. After visiting Queens, Mulanjie, Thyolo, Chiradzulu, and Zomba and conducting some further research, our team quickly came to the conclusion that the calf, rather than the foot is the best place to target when attempting to prevent blood clots in bedridden patients. Because the major veins of one’s lower extremities are initially supplied by the calf, blood clots tend to originate in the calf before traveling elsewhere, making the calf a more effective target. Additionally, calf compression has been shown to be more hemodynamically effective (this includes measurements of peak velocity, refill time, and cycle volume) than foot compression and is the most widely supported form of IPCDs.

Of course, targeting the calf rather than the foot means that various technical specifications of the device must be altered. This includes the applied pressure, the duty cycle, and the cuffs themselves. While implementing these changes will greatly increase the effectiveness of the device, our team will certainly face some significant engineering challenges carrying out these modifications. However, we have a good grasp on how the device needs to be changed and how those changes should be made. In general, there are 3 major changes that need to take place in order to efficiently switch to calf compression:

1. The cuff should undergo the following modifications
   1. The cuff should be a rectangle that’s 60 x 30 cm
   2. The airbladder should be a rectangle that’s 30 x 27 cm
   3. Rather than backpack straps, the cuffs should be secured with Velcro and simple buckles
   4. The material used to make the cuffs should remain the same

(Current Foot Cuff)

1. Rather than an applied pressure of 120 mmHg, the device should achieve an applied pressure of 40 mmHg
2. The duty cycle should be altered such that
   1. The inflation time is slower (5-10 seconds)
   2. The deflation time is slower (5-10 seconds)
   3. The cuff stays inflated for a period of time (“hold time” of 5 seconds)
   4. There’s less time between each cycle
   5. Both the left and the right cuff inflate and deflate in sync rather than inflating and deflating on alternating cycles

(Current Duty Cycle)

While modifying the cuffs is a relatively straightforward task, modifying the applied pressure and the duty cycle will be slightly trickier tasks. As of right now, there is no way to control how fast the air flows out of the air tank and into the cuffs. Slowing down the inflation and deflation of the cuffs will require intense modification of the software that controls the device. Essentially, because the valve that allows air to be released into the cuffs is controlled by a digital signal, we’ll have to use pulse width modulation to get analog results with digital means.

Additionally, because the right and left cuffs must inflate and deflate in sync, the airflow system must be modified such that air can flow into both cuffs at the same time. This may require using a larger pump so that both air tanks can be filled with air simultaneously. Alternatively, we could modify the airflow stream such that both cuffs are filled with air from the same air tank. However, this may not be ideal, as it requires a larger air tank and an air pump with a higher flowrate.

I’m sure our team will have a great time figuring out the best way to solve these issues in the upcoming weeks!

The engineering design process teaches all of us the method we must follow in order to guarantee flushed out designs and viable products. It’s first step is stressed as the most important: brainstorming. Solid, effective, and meaningful brainstorming is essential as it expedites the design process by eliminating the need to completely start over. However, an important step to the process is recognizing when one can move on.

My teammate and I learned this important notion during the brainstorming week. We only spent two days brainstorming while everyone else spent a whole week in the process. During our brainstorming session we came up with different methods of manufacturing cup seals from PTFE sheets. PTFE otherwise known as Teflon is commonly used as the non-stick coating on pans. However, in compressors its self- lubricating properties allow it to be used as a cup seal.

We understood that we must create a method for manufacturing these seals in an easily understandable, sustainable, low-cost, and precise manner. Thus we brainstormed the following methods: a puncher machine, a crank powered “cookie-cutter”, scoring machine, a hand held “cookie-cutter”, CNC machine, and laser cutting.

After scoring the ideas in a Pew scoring matrix we concluded that the most practical ideas would be the puncher machining, laser cutting, and CNC machining because of their ability to be precise, semi- autonomous, and sustainable.

After having a talk with Matt we reevaluated our original purpose and decided that immediate testing was the key component of our project. So during the second day of brainstorming we started making prototypes of the cup seals using the laser cutter and began testing the very next day. We perviously learned that vibration, dry friction (from dust), and extensive use were the main causes for cup seal degradation. And that replacement of worn out cup seals is not a procedure carried out by hospitals in Malawi. These worn out cup seals aren’t able to form a proper seal thereby air leaks (the compressor’s output decrease) which decreases the pressure and air concentration. So we deduced that we should measure the pressure output and oxygen concentration of a O2 machine with our cup seals to track their performance.

To make sure that we didn’t damage the machine’s sieve beds (the containers filled with the nitrogen stripper which is connected directly to the compressor output valve) we did our testing on old O2 machines with faulty sieve beds. As a result to using faulty sieve beds we only measured the pressure output hoping for a pressure output range of 15-30 psi. Our prototype performed very well.

Brainstorming is essential and its impact is definitely well-stated; however, sometimes it may play a small role in your project.

The celebrations during the summer are always my favorite part of the season. Because the days are longer and school is out, summer, especially when I was younger with less responsibilities, was always spent around the barbecue in our backyard with friends and family. The past week was double the fun since both American Independence Day (July 4th) and Malawian Independence Day (July 6th) happened around the same time. And if that isn’t enough Madonna visited the week after to open up a new hospital! However in spite of all the excitement our project still continues; this week we will have another round of tests which will hopefully yield useful results, exciting in a different way.

With the data collected from the previous preliminary testing conducted on the different filtering materials as well as determining possible entry points of dust into the machine we have developed 3 prototypes with attempt to address the problem. The first is simply placing the filter behind the external sponge. While cheap and easy to implement it may not be the best solution since it may block too much air which is used to cool the machine. The second design is a baffled filter box. The folded filter paper has a larger surface area so more air is able to pass through. Finally the last design is an active filter box which uses an external fan to pull air across filtering material into the machine. This design can channel air exclusively from a single area and prevent dust from entering through other vents due to a zone of positive pressure being formed inside the machine as well as force more air into the machine which can be used for cooling. These designs will soon be tested on one of the concentrators hopefully providing good data on which prototype is most effective at allowing proper airflow; internal compressor temperature will be measured after 4 hours of filter testing for each prototype. The results will supplement the week long tests conducted at the end of the week on which proposed solution allows the least amount of dust to enter.

But even with all the fun and festivities around us, as with most great celebrations, they will eventually end. with only a few weeks left we are nearing the end of the testing and the internship. But even though I am feeling homesick right now I have a feeling once I go I’ll be sad to leave this picturesque nation behind. On one hand, I will be happy to spend time with family and friends in the states again, but on the other hand, I’ll miss the new friends I’ve made here at the Polytechnic, Kabula Lodge (the place we are staying), and even the fellow Rice interns which I’ve had the pleasure to get to know better. I’ll just have to make the most of the remaining weeks, making memories and hopefully producing a prototype that will outlive my time here.

Something I have noticed about myself is that I have a very singular focus; I tend to do a lot is devote all my time and focus to analyzing either an idea or an activity, unable to pay attention to other things while this is happening. When I was little I would frequently read books in one sitting, unable to put them down even to go to bed. Although I’ve gotten better at multitasking, this still tends to happen at times. This is sometimes useful; I have been very grateful the night before an organic chemistry exam or before an important presentation. But by redirecting your focus in this way some pretty unanticipated things can happen, often times helping me refocus on the big picture rather than the nitty gritty.

This has been especially applicable for the project, I am currently working on. If you told me right after my previous post that I would continue my initial project, skepticism would be an understatement. I had already begun to internalize what the technicians said and began to focus on other aspects of the concentrator: airflow, ventilation, etc. But what I had ignored was the big picture. Compressors break unnaturally early and something must be causing it. With helpful input from George, a veteran technician employed by the Rice CPAP office for the nationwide implementation of Pumani bCPAP (a low cost medical respiratory device designed to help babies begin breathing), and Matt Petney, the Polytechnic Design Studio director, I was able to refocus on the issue at hand and re-examine the different components that may fail in order to find the problem.

After extensive investigations my partner and I found that the external gross particle filter on oxygen concentrators is wholly inadequate at stopping dust from entering. We ran a battery of tests on the concentrator for this including measuring internal temperatures at different time points, placing concentrators against walls to see if ventilation may be an issue, taking out the compressor and mapping the airflow route of the cooling and compressed air, and of course the extent filtering in the machine. Most of these tests returned normal results except for the filter tests.

In order to test filtering, we designed a airflow detection system and a dust retention system in order to see what types of filtering materials would both allow a lot of airflow into the device and also retain to most amount of dust that can be picked up from the surrounding environment. The airflow test device consisted of two tubes that were separated by a filter with air being pushed across with a fan that could be monitored by a air pressure sensor hooked up to a Arduino processing unit. The dust tension device used a fan in reverse to pull dust from one side of a filter to the other side into a collection container. As expected, the gross particle filter preformed the best at allowing airflow but just as George mentioned it did nothing to stop the dust. (The fan didn’t even be needed to turn on to see this happen). Other materials we had previously acquired were tested as well in order to cover the bases and determine which filter material would be the best should we chose to replace the filter.

But in the meantime we have been sorting through different ideas we had for other ways to solve the problem. After a sorting the weighing the different solutions, we came up with creating a an external filter box which can either passively take in air or actively force air in through the help of another fan attached the the concentrator. Currently we are in the process of making prototypes of these devices to determine which system is better at solving the problem. We will test them over the course of this week; hopefully they will yield positive results.

Besides conceptually reconsidering my project problem statement, this week has also forced me to refocus in a much more concrete (and unfortunate) way. Because of this project, I have been spending my free time trying to learn how to use Inkscape, a free graphics design software, to make containers for the filters as well as brush up on other design software. My computer literally had to die in order to stop me from using Inkscape. Currently I’m typing this blog post on my phone which is an experience I would definitely not recommend. However maybe this unexpected incident will help me rethink the big picture instead of pigeonholing me into designing a filter box. While this is our current avenue of approach we do have other ideas that potentially may be better. Who knows?

The more time I spend in Malawi the more I feel a sense of urgency to return to the land of my birth. This sense of home I feel here reminds me of the few memories I remember of visiting my country. My time so far has taught me that the most significant impacts are made locally. It’s through locally created and designed solutions that projects and advancements become sustainable achievements. I think that’s why this program is so meaningful. Direct impact has always been a priority. Two weeks ago we spent three days traveling to the district hospitals of Southern Malawi.

The first city I visited was Mulanje, home to Malawi’s tallest mountain. The scenic drive did little to prepare us for the spectacular view of the Mt. Mulanje which was situated right behind the hospital.

Being a district hospital it was much smaller than Queen Elizabeth- the hospital near the Polytechnic University. Walking around we learned that the sole X-ray for all the southern hospitals recently broke beyond repair, that long queues are a daily phoneme, nurses are plagued by severe understaffing, and that neonatal devices break constantly. It was eyeopening to see the level of improvising that nurses and doctors needed to do. Having spoken to the nurses and learned as much as we could on their cleaning methods, we spoke with the PAM (Physical Asset Management) director as he knew more on how these medical machines function and break. At Mulanje dust is an issue for their oxygen concentrators; however, maintenance procedures are in place to help alleviate the breakdown of these machines. Their preventative measures include weekly cleaning of filters and monthly internally cleaning of dust. Furthermore we gained insight into how dust may play a role in the accelerated degradation of the cup seals and the need for locally manufactured cup seals. They welcomed the idea of locally made cup seals as this would allow them to quarterly replace worn out seals thereby prolonging the lifespan of the concentrators. I walked away from Mulanje with a sense of realistic hope, even though there existed a myriad of systematic issues locally made spare parts could make a significant dent.
(insert pic)

After Mulanje we drove past acres and acres of tea plantations to arrive at Thyolo District hospital. You could tell this was a recently constructed hospital. The layout of the wards, the well kept garden and sitting areas contrasted the congested and convoluted layout of Mulanje and Queens. Architecture can really make a difference! We went about the same routine; visiting the nurses first and then speaking with PAM. The oxygen concentrators were placed near the ground and right against a wall. This placement contributes to their shorten lifespan by exacerbating the effects of overheating and humidity destroying the filter mineral. In addition, nurses only externally clean the concentrators weekly.

And with no internal dust removal the accumulation of dust practically destroys the machine. Thyolo like all the hospital we visited no longer has extra compressor cup seals and sole rely on the Rice CPAP team to give them some. Therefore, they use the concentrators until the cup seals’ breaking point- the point at which they are no longer is able to maintain the necessary pressure to concentrate oxygen. PAM’s response to our questions was essentially identical to previous responses.

The following day we visited two more sites- Chiradzulu and Zomba- and received relatively the same feedback. Since Zomba hospital is a central hospital (meaning its much bigger) we decided to spend our time there exclusively with PAM.

We were greeted to a large “graveyard” of broken oxygen concentrators. Whenever a district hospital can’t fix a machine themselves they send it to a central hospital like Zomba; however, for the most part these central hospitals also lack the spare parts to actually fix these machines so they end up stacked up on shelves. Since Zomba has one of the largest PAM facilities we decided to berate them with our questions which concerned the durability, material specification of cup seals, and hospital prevention procedures. We learned that dry friction and overheating were the main causes of degradation, that spare cup seals only come in kit that includes other compressor parts making it expensive to order, and that there is a real need for locally made seals. Since many devices- autoclaves, ventilators, and air compressors- use cup seals there is a potential for a commercially viable product. Traveling has really energized and moved me in a exciting way. Day in and day out the Malawian interns and I exchanged knowledge, hopes, and experiences. In a sense we have opened each others’ worldview. We inspire each other to use this internship and what we have learned as a starting point to do so much more for the global community. And with all that I have learned here, I know that one day I will return to Eritrea and use my experiences to do my part.

Last week we came up with a design criteria and it was evaluated using pairwise comparison chart. 

We brainstormed 8 designs of protecting compressor from dust which were evaluated using Pugh scoring matrix and the selected design criteria.

Three designs were selected to be tested then implemented if they meet the desired requirement:

·         Improved external filter

·         Filter box to be attached outside the concentrator

·         Attach a positive pressure fan

A 240v Ac positive pressure fan was connected to the concentrator and tested to determine if it can control the entry of air into the concentrator thereby reducing entry of dust.  So far the results are positive.

Shown above are some of the filtering materials which were bought last week from Cash Build and Shoprite. They are to be used to improve external filter and to build a filter box with the aim of controlling entry of dust into the concentrator.

We designed an air flow test experiment consisting of a fan, two cylindrical plates, Arduino, pressure sensor, connecting pipes, in order to test the porosity of each filter. Each filtering material was placed in between the cylindrical plates to test how much air it could allow to flow. The reading of pressure was printed on the computer after the fan was switched on. The control filter (black) gave a reading of 49 units (lots pf airflow), the cleaning sponge (pink and green) indicated 51 units (moderate airflow) and scotch brite (green) and filter roll allowed 52 units (moderate airflow).