Blue Team!
Mockups!
Open in New Window - All Presentation Slides (.pdf)
Presentation Video
Demo Video
Concepts
Drowning Alert
A wearable wristband that detects when a child is drowning and uses light and sound to alert people nearby.
View Product Contract in New Window
Contributors
Joseph Ntaimo: Sourcing Parts, Hardware, Software, Testing, Talking to Users, CoordinatingJosh Sohn: Hardware, Testing, Presentation, Talking to Users, Drowning ResearchCarly Long: Hardware, Testing, Product Contract, Talking to Users, Drowning ResearchDanielle Allison: Talking to Users, User and Market ResearchMulan Jiang: Hardware, Testing, Presentation, Talking to Users, Coordinating
Reviewer Feedback
Ellen Roche
Concept
I wanted to point out another group of people that could benefit from this device, especially if manually triggered and if able to send a message to a friend/lifeguard etc. I used to do open water swimming and would swim in the sea/lake early in the morning but would always swim with a friend in case somebody got a cramp or struggled or something happened in the water. This was sometimes difficult and if one person wanted to go back both people had to. I think if you could use your device as a wearable that could trigger an alert to a friend on land or in the water somewhere else it would give incredible peace of mind and allow swimmers to train even if they didn’t have another swimmer at the same pace etc. There are also inflatables that can be worn on a belt and if that could be triggered on demand it might provide additional safety for the swimmer (or kid as per the original target market)
Analysis
I think there is a risk of sensing SpO2 and no movement as I worry this would be quite late to detect when somebody is in trouble. I feel like a manual trigger that could be deployed by the child in trouble and that alerts a parent or carer might be useful - I guess lots to consider if you move ahead with this exciting idea
Execution
Nice demo underwater!
April Anlage
Concept
I think you could further clarify your user - you mentioned that about 1/3 of drownings occur in pools with lifeguards, leaving me to think the other 2/3 occur at home or other scenarios where a professional might not be on hand. To me, these seem like very different markets with different needs.
Analysis
I still think your biggest risk is actually detecting drowning. Obviously this is difficult to simulate - the 3 sensor model you proposed here is a step up over what you had in sketch model, but the actual sensing is still very nebulous. For example, how do you quantify an SpO2 level that indicates drowning? Is there data on this? Is SpO2 level affected by age, activity level (i.e. kids splashing around), hydration status, etc etc all the thing you can't control for? What sort of SpO2 level indicates danger but not too much danger? Finding that sweet spot (and the confidence level associated with it) is a significant challenge. These questions also needs to be quantified for the other two sensors. There is also the additional challenge (that you seemed to have explored a bit more) of actually alerting someone. Again, this will depend a lot on your chosen market (see above). Sound seems unlikely to work. Can you test how well your light alert (or other alert) works to get someone's attention?
Execution
You seem to have learned that sound will be a difficult detection method, which is a valuable learning. However, it would have been good to see you pivot to other options. Additionally, much needs to be quantified with your sensors as noted above.
Adam Frick
Concept
I like where this pivoted from the sketch model, where the product is envisioned for use in public spaces.
Analysis
Emitting an audiovisual alert from the bracelet will prove a challenge for this environment, unfortunately. An alternative: a lifeguard distributes the bracelets to all swimmers in your target age group, configured to alert the lifeguard directly via bluetooth or RF. I may have mentioned this in the sketch model feedback already, but I am skeptical about the metric of motion to determine if a user is drowning. Drowning with water turbulence (such as in your wave pool video), or on a shallow end of a pool may result in false negatives with your system.
Execution
Moving forward, I raise some caution about the accuracy of using blood oxygen sensors. The measurement may be affected by skin profile and snugness on the wrist. This would also prove a challenge in testing (I don't recommend self-induced hypoxia). If all other detection mechanisms fail, I think the user should be able to trigger an alert manually also.
Kyra Post
Concept
The video you shared demonstrates how drowning can easily go unnoticed, and makes a clear case for why it would be helpful to have an assistive device to better draw attention. I was glad to hear you have spoken with both parents and lifeguards, but another important user is the child themself. The device only works if the child wears it in the water-- what if they take it off and leave it on the side of the pool? Young kids might find a device bulky or uncomfortable. Preteens/teens might be embarrassed to wear it or think they don't need it anymore. Continue speaking with parents (since it sounds like they will be your buyer) and lifeguards (since it needs to provide signals that are noticeable and meaningful to them), but also consider ways to make sure it is child friendly or ensure it cannot be easily removed without a parent's help.
Analysis
You've proven that you can DETECT drowning conditions as you have defined them, but I don't think you've found a solution that properly NOTIFIES adults of a drowning child. This is a major risk. It's not just a matter of the light being "visible" when you are looking for it (what if it's too sunny to see, or there are people blocking your view to the wristband?); it should be so obnoxious that everyone in and around the pool knows something is wrong (think of other emergency notification systems: fire alarms, ambulance sirens, etc). As you showed, the sound emitted directly from the device is not loud above the surface. Consider other ways to transmit the alert-- can the device trigger a loud alarm on the parent's phone? In a separate poolside receiver (again, big red siren?)? Could you have paired parent-child wristbands so when the child is drowning, the parents wrist alarms/vibrates? Without a strong alert solution, this concept may be dead in the water (sorry I had to)
Execution
I believe your mockup can detect depth which is one key indicator of drowning, but I'm not clear on the technical advances since your sketch model besides adding a screen. Have you done any testing with blood oxygen sensors as you mentioned that could be the first signal of drowning? If you haven't already, be sure to test this in an outdoor pool on a sunny day-- the plastic tub is good for technical validation but not super representative of a real use case. Since the sound from the wristband didn't work as you might have hoped, I think your key risk if you choose to carry this forward is how you will effectively alert adults of the drowning child. Be sure you have a clear idea of how you might solve that problem before deciding to continue.
Jordan Tappa
Concept
The problem is super clear and I strongly commend your video example! I worry about implementation of your device as is. As you've discovered, trying to emit sound outside of water is neigh impossible, and your blinking LED is simply more visual noise in a chaotic pool setting (and invisible in daylight). Consider back to the core requirements of what a device like this needs to do to function. Be triggered when drowning is occurring, tell a lifeguard or competent swimmer that you need help, and/or save you. Ask yourselves, what parent is worried enough to buy this for their kid, and yet not enough to make them wear a life jacket.
Analysis
I think you learned a few valuable lessons about waterproofing and sensing underwater. The flow chart was also a nice touch. However I'm not convinced that this is sufficient to distinguish between swimming and drowning. For example, floating on your back and floating on your stomach are almost the same, however in one of those the swimmer has died. And if the difference is the pulse oximeter, would there not be too big a delay between when the device triggers, to when a person recognizes that drowning is occurring, to the time they can actually do something.
Execution
Consider alternative methods of alerting lifeguards and/or neighboring swimmers. Lights and sound are - from your findings - insufficient for providing notice. Consider potentially radioing to a life guard, or connecting with a drown alert app on a swimmers smart watch, and other out of the box ways to notify distress.
Ben Nahill
Concept
Cool problem! The concept and user seem clear, and it seems potentially viable!
Analysis
Specifications seemed roughly sound. You've hopefully found whether an underwater alarm can be audible enough.
Execution
Pressure sensing is clearly pretty good. Motion detection is relatively straightforward. The alert aspect seems like the area that needs the most work. Even if audible, you need something that will draw attention in a noisy environment. Even above water, that would be a challenge for a wearable device. Consider coupling with a louder edge-of-pool device that could aid in the alert process. There are waterproof ultrasonic buzzers that could be directly exposed to the water and could produce signals which carry well in the water.
Juergen Schoenstein
Concept
The brief video in your presentation made a very strong case for a drowning alert system, no doubt. But it also showed why this is a very hard problem to solve: In a crowded swimming pool, with all the commotion and reflections and distractions, identifying a single individual - ANY single individual, drowning or not - seem daunting. and with the telltale signals for drowning being so subtle, it seems unsurprising that it can go unnoticed until it is too late. So, from the standpoint of value that a solution could provide, a drowning alert is a strong concept.
Analysis
That said, you also learned (or at least I hope you learned) that the technical solutions you had in mind might not be suitable to the task: the light, as a visual signal, will be essentially invisible on the surface (again, lots of reflections, wave action, other people being quite agitated). Finding that you can spot the light from a distance, once you know where to look and have no other distractions - which seems what you tested - is nearly irrelevant; would you be able to see it under the conditions shown in your video that established the need and the situation in which you want your idea to work? I did not see a compelling answer to that. The sound alert is even more problematic, since physics create a reflective barrier for sound at the boundary between water and air (the sound does not leave the water, but gets reflected). Unless you install underwater microphones, the alarm will not be detectable, no matter how loud.
Execution
Showing that a sensor can detect water pressure and trigger a signal is certainly relevant to your project idea - but that was not really surprising (in other words: you knew the answer, and it was not critical to verify that again). I was underwhelmed, and would have liked to see maybe some other solutions and/or more pertinent data (like: how discoverable is the light signal under chaotic conditions).
Georgia Van de Zande
Concept
It sounds like you have a good use case of a crowded pool and careful parents. There is some more you can do this weekend to further clarify the user. The kid is definitely wearing the wristband, but who bought it? The parent or the pool (and then the parent rents it)? The lifeguard needs to know this device is out there, or they’re not going to know what a flashing light means. Talk to summer lifeguards, not just the ones at MIT. These are very different pool scenarios, so make sure you are talking to ones that match the “lots of kids playing outside in the summer” profile.
Analysis
Yes, the light is visible from 25m away, but it’s likely not noticeable, especially when the sun is reflecting off the water surface too. How else can you change the product architecture to make something that is noticeable? I am thinking you can install something permanent under the surface of the water. That gets a signal from your bracelet, and it is connected to a siren or light above water that the lifeguard knows to look out for. Of course, this goes to my questions before about who buys this product.
Execution
You built a model that could sense how deep it was in the water. A start, but one small piece of this device. You still have some tougher questions remaining. What is noticeable in a busy pool? How to sense blood oxygen? These two come to mind, but there might be more. Before your decision, think more about what are the real risks, and try to address those.
Adaptable Pregnancy Support Belt
Flexibelly is a pregnancy support belt that supports pregnant individuals’ backs and can be easily adjusted to tailor compression to the position the person is in.
View Product Contract in New Window
Contributors
Andrea Moncada: user interview, model ideation and assembly, presentation slides, product contract, sensor testing/ recordingEddie Barrios: model ideation, user interview, product contract, sourcing for sensors, sensor testingKaira Samuel: User interview, presentation slides, model ideation, product contract, sensor testingChristina Patterson: Presenting, Coordinating, sensor testing, model ideation and assembly, product contractZoe Wu: Sourcing for parts, user interview, model ideation, product contract, presentation slides, sensor testing
Reviewer Feedback
Ellen Roche
Concept
The user is clear - it would be good to establish whether the belt could be adjustable throughout the sensing and whether there could be an inbuilt simple sensor to identify the level of pressure being applied (4-15mmHg is recommended) - there’s a company called Feeltect you can look up who do a nice sensor like that for compression bandages (it’s called tight alright! ) . Rob Podoloff the red team instructor can likely help here too as he has worked with a lot of flexible sensors if this is something you’re interested in. If you go without electronics I think a very easy method of adjustment in an area that wouldn’t feel uncomfortable would be key. Sometimes the sides of the expanding belly are super tender and itchy so less hardware there is probably better!
Analysis
Maybe some more analysis into compression levels would be helpful
Execution
Nice mock-up but see comments above about location of hardware
April Anlage
Concept
Are you adding significant enough value compared to existing products? Are there other value adds you can incorporate?
Analysis
I see in your product specs "relieve back discomfort" which you then link to the Young's modulus of the material - this seems unrelated to me. In order to relieve discomfort, it seems to be you would need to measure something physiological (lift of belly? compression? pain rating?). It's worthwhile to do this exercise with all of your specifications to make sure they are specifying what you intend. Additionally, you proposed that back pain in pregnancy was a result of the changing center of mass of a pregnant person - how sure are you that this is the only (or even most significant) contributor? A quick Google search suggests that stretching ligaments is a common cause of back pain in pregnancy. It's worth looking into the complicated physiology of pregnancy such that your solution addresses the challenge as thoroughly as possible.
Execution
See comments above - there are many things you could additionally measure here in order to demonstrate the functionality of your product.
Adam Frick
Concept
A broad problem space, much potential for users who may not even be aware of maternity support devices!
Analysis
How did you determine that your product successfully provides back support, notwithstanding the similar functionality of competitors? You identify 5 inch lateral expansion is possible - does this figure also come from meeting/exceeding design specifications of existing products? My concern is that if your product behaves "too" similarly to competitors, then it may be difficult to stand out, despite the quick cord lock.
Execution
User testing seems critical moving forward, including feedback on your product and the products it is up against.
Kyra Post
Concept
It sounds like you understand the existing pregnancy belt landscape and as a result you have developed a clear and differentiated value proposition.
Analysis
You have highlighted features that would make your product better than existing ones, but are you sure they would fully meet the need? For example, you mentioned 5" expansion is better than what's on the market, but is that truly the maximum expansion a pregnant person's belly would experience between sitting/standing? What about standing in early pregnancy (smallest) vs sitting close to your due date (largest)? Make sure you define your product contract so your solution will be awesome, not just better than the sub-par options out there. I think the 1-handed adjustment and no need to reach around is a key insight-- you can test this a bit on your own with the fake pregnancy belly to get a sense of where you can reach. Bonus if you can find 1 location that works for both righties and lefties! Be sure to also test this on real pregnant people. I am not clear on how you are measuring force relieved/compression by the belt but regardless of what you can measure numerically, it is important to figure out how those values correlate to user comfort and relief.
Execution
This is a simple mechanism but seems to work easily and is a step forward from the sketch model. If you haven't already, testing on real pregnant people will likely give you great input for future iterations. I didn't see the model in person so can't comment on how comfortable the material/form feels, but this will definitely be important to test with real user feedback, adjust, and test again...and again... and again... One question I'm left with on size-- it is obviously adjustable in girth but what about height? I imagine a band that is too tall would be quite uncomfortable for petite people (dig in in all the wrong places), while a band that's too short for a tall person could mean not fully supporting the relevant part of the back. You likely only need one model for an alpha prototype, but it's something to keep in mind as you consider how to adjust the fit to be just right!
Jordan Tappa
Concept
The value of a 'better' pregnancy belt seems strong. Right now your team is focusing on the waistline adjustment, and the extra inches gained. Are there other inefficiencies and pain points that these devices cause for pregnant women that can further widen your product opportunity? Are there passive systems that still provide support and expansion without the mother adjusting the belt? And why might someone when looking at the wide array of pregnancy belts available, buy yours instead?
Analysis
I was curious about the support you're providing on the user's body. Would have been nice to see some examples of the force that competitors use to support and the force your belt exerts. I also suspect as your loosen and lighten the elastic section that the force is likely to change. It may be that simply having that adjustable section allows users to tighten to their specification and comfort.
Execution
A relatively simple model but a solid step in the right direction, would have enjoyed seeing explorations in other methods of expansion or other ways to allow pregnant users to sit more comfortably.
Ben Nahill
Concept
Very cool idea. From the little I know, this is a real problem for some mothers-to-be and this seems like a potential solution.
Analysis
Seems reasonable.
Execution
The adjustment mechanism is simple and quite effective. Your concern about friction could perhaps be addressed by replicating the adjustment on both sides. For ease-of-use, there are cordlocks I've seen on backpacks and mountaineering boots which tether to the material so that they can be snugged up and released with a single-handed pull, rather than a second hand fiddling with the cordlock itself. I can't find a picture at the moment but I'm sure you've used them before. Definitely try to get some real user testing ASAP.
Juergen Schoenstein
Concept
It seems credible that the need for your idea/solution to provide support and relief for pregnant women exists, since similar products are already in the market and (apparently) widely used. And better solutions are certainly a valuable contribution to the quality of life for expecting mothers.
Analysis
I will lump this answer together with my reflections on the execution of your model, since the purpose of testing was mainly qualitative (which makes sense, since comfort - which seems to be your main goal - is essentially a qualitative term). I found your demonstration of your tightening/loosening mechanism compelling, and with some tweaks, it seems absolutely feasible. But keep in mind that your goal is not merely to mechanically solve a tightening/loosening problem: Your claim is that you can do that and increase the comfort of the wearer. And that requires a lot of fine-tuning and, yes, user testing. It is one thing to make wearables functional - it is quite more challenging to make them comfortable (just think how many pairs of shoes you may have had to try on to find one that is actually comfortable for you). Your mock-up is a good start, but I would suggest to a) explore more adjustment mechanisms, and b) fine-tune materials and geometries to ensure that the comfort of the wearer is maximized.
Execution
See above
Georgia Van de Zande
Concept
It sounds like you’ve done good benchmarking against existing products and have found a niche in the ultra comfortable, but meant for lounging around the house belt. That’s good. You can still focus on form to make the device attractive, but you don’t want to make any concessions when it comes to comfort. Exploring different ways to tighten or loosen the belt would be good, and thinking about new placements could help you find the most comfortable.
Analysis
It would be good to measure pressure that the belt causes, and compare that to existing products. As you noted, the stuffed belly is softer than a pregnant belly. Moving forward, you’ll also want to test with a pregnant mother. Do you have someone in mind who is willing to help? She can also tell you if the device is providing enough support compared to the existing products.
Execution
Nice job getting used to working with soft goods (one of the instructors said you were just learning).
Anti Panic Attack Vest
A vest that applies a calming pressure to the user at the press of a button to help prevent or mitigate panic attacks.
View Product Contract in New Window
Contributors
Jonhenry Poss: Research, CAD, 3D printing, parts sourcingPablo Alejo-Aguirre: Interviews, assemblyLuis Ibarra: Interviews, assembly, parts sourcingSophie Longawa: Research, assembly, sensors and electronicsDaniel Amaya: Interviews, sewing, assembly
Reviewer Feedback
Ellen Roche
Concept
Clear user, nice idea I’m wondering if you could power with a canister of CO2 or something instead of carrying around a pump and hardware
Analysis
Maybe some more analysis on the level of pressure application would be good. If you move forward please let us know! We have lots of experience making McKibbens and other types of soft actuators and can definitely help!
Execution
Nice mock-up, good textile integration - orientation of McKibbens and powering mode could be characterised and optimised further
April Anlage
Concept
No feedback provided
Analysis
Your most critical user need is completely missing - isn't your product designed to mitigate panic attacks? To that end, what is your specification to determine if the vest is working? Have you been able to quantify what amount of pressure needed? Where does the pressure need to be directed? There is a significant physiological component here that needs to be quantified.
Execution
It sounded like the technology you chose didn't work as well as you had hoped - that's ok, but models like mockup are an opportunity to explore other mechanisms - what are your other ideas for applying this pressure? And again, you need an objective measure to determine how well your product and chosen mechanism is working.
Adam Frick
Concept
Does deep pressure therapy solely address panic attacks? I wonder if its sensation is a general means to relax the upper body, or if it could also be helpful for people with varying forms of neurodivergence.
Analysis
Battery life and safety were mentioned up in the slides but I did not see these addressed. What concerns do you have about battery safety (power consumption, exposure to high temperatures and moisture)? What is the expected power draw with the spec'd pneumatic pump? Since the vest is to be worn between clothing, is sweat resistance a concern without contact with the skin? I am concerned about the lack of spec for the actuation time - addressing panic attacks seem time-sensitive, especially in public. This, noise level, and energy usage seemingly all affect each other
Execution
It would have been nice to see a working example of the pneumatic muscles outside of the vest form factor. Discreetness is a large factor in compelling users to wear one in public, so this would have been nice to demonstrate by putting an article of clothing over the vest (even in its resting state). Despite the mockup's pump being larger than the one you envisioned, where was it designed to be stored on the vest/body?
Kyra Post
Concept
I can see how a real human hug would be comforting, but I'm not 100% convinced that a self-actuated vest would have the same effect (is the pressure from hugs what has the calming effect, or is it the presence of another person? There may be literature on this I'm unaware of). I'm also not sure how often one could wear a full vest like this-- would I need to wear baggy clothing all the time to conceal it, or be constantly explaining to people what this puffy vest is? Even 0.5" all the way around my body would probably be too much added thickness to wear under my normal shirts. Have you done any testing with a user while they are experiencing a panic attack (I realize this would be impossible to schedule but maybe you can find someone willing to try it for a few days)? Overall, I believe in the problem, but without evidence of the effectiveness of this kind of mechanism I am a little skeptical of the product.
Analysis
Your user needs are clear but I am concerned about some of the specifications (see above re: low profile). 50 dB is like a slightly-quieter-than-average window air conditioner-- probably ok in your living room, but would definitely be noticeable if someone next to you in lecture started making that much noise. Make sure you have tested/experienced that amount of noise before setting it as your upper spec. Is there another way to actuate this that would be nearly silent, with the majority of the noise coming from body movement or rustling your clothes? See comments from other reviewers about mechanical actuation
Execution
Great looking sewing (majorly more stylish and believable that someone would actually wear it so good progression from the sketch model). I couldn't see that it was increasing pressure, which is probably a good thing for it to stay discreet. But for believability's sake or for better pressure control, you might want to add some pressure sensors to make sure it is targeting the correct areas of the body (ie not squeezing too tight on soft areas or bones and not enough on more muscular areas or whatever it is that should be targeted)
Jordan Tappa
Concept
While panic attack mitigation is a strong need, I find a vest a curious form factor for this product. It seems as though you mean for users to wear this product throughout the day and trigger it if they need the 'hug-like feeling', however I end up taking my hoodie off most days when I get inside and don't see users comfortably wearing this the majority of their day, everyday they need it (plus finding time to wash it). And if this is meant to be stored in their bag and brought out when needed, then it wouldn't need to be a vest as much as a constricting sash or blanket, etc. Also curious if this needs to be an electronic device at all, and if instead it could be mechanically operated by the user (pulling straps to actuate the hug as a mundane, repetitive activity that might calm them as well) or simply an elastic compression vest which is what the actuated 'muscles' are doing anyways.
Analysis
Unclear how much force the actual model was exerting, a 'hug-like feeling' kept being used as a comparison, but would be nice to have a frame of reference for exactly what the vest is doing. It was unclear whether the tug was quite weak or very strong and which of those effects would be best for a wide variety of users. I would be a strong suggester of switching to something the user can pull to tighten and loosen and effectively hug themselves.
Execution
The vest looked sleek, I commend your sewing work. A bit too much uncertainty around the pneumatic 'muscles' for me. For a device that is 'discrete' pneumatic seems too loud and bulky for this use case.
Ben Nahill
Concept
Also a good problem, albeit one I'm less familiar with.
Analysis
Intuitively this seems a bit obtrusive but then it may be worth it for some.
Execution
Pneumatic seems like a cumbersome way to go. If you haven't tried a corset-like mechanism I could see that being much more compact (in both the garment and any external components) and energy efficient. If you must, pumps and associated power systems are more complex and heavier than CO2, but also may last longer, especially if the system isn't perfectly sealed. Definitely consider all of your options, as pneumatic is challenging and complex. If fit may vary, perhaps use pressure sensors (even just force-sensitive resistors) in key areas to determine when the target pressure has been reached.
Juergen Schoenstein
Concept
I am not sure that I had a chance to understand the value that your idea will provide. I am somewhat familiar with anxiety attacks (a person very dear to me is struggling with those, and - at least for some time - was actually on medication to mitigate those attacks), but I have not seen the benefits of hugging (actually, that person gets even more agitated when "constrained" by hugging - but that is a "data point" of one anecdotal experience, admittedly). I would have liked to see more analysis going in already: What does literature tell you about which pressures at which locations for which duration can mitigate what kind of anxiety attacks to what degree (supported with data)? Most of the evidence that you invoked seems to be based on use of this approach with animals (dogs, I suppose), but how solid is the evidence that the same effect can be achieved for humans? What do we know about the aspect that humans would have to self-administer this "treatment" (which might undercut a potential emotional effect of feeling cared for, for example)? I see a high barrier to making this concept convincing for a broader audience, and "trust our word, it will work" might not reach high enough.
Analysis
I would have liked to seen more data - on pressure exerted on the body, on heart rate when applied (since the claim seems to be that the effect is entirely physiological, it should be observable in some biometrics, even when the test person is not prone to anxiety attacks). I am concerned that truly testing this concept (and, even more so, a product built around this concept) will exceed whatever resources you may find over the next few weeks. And my comment on comfort (see Flexibelly feedback) applies here just as much: To make sure that this vest can be worn comfortably, you will have to solve - and test! - much more than just the technical challenges to make it work mechanically.
Execution
I was impressed by the craftsmanship of your model - I know how to use a sewing machine to create a fairly straight seam, but this vest did require a lot more detailed work, and it seemed quite well executed. But it was hard for me to see the vest "in action" - we had to take your teammate's word for it that something was happening.
Georgia Van de Zande
Concept
It sounds like you have found a new way to provide compression, compared to existing products. The concept to use the finger traps is interesting. I don’t know a lot about soft robotics, but it would be good to talk to the experts around MIT who do. How do they actuate them?
Analysis
It would be good to measure pressure on the chest, and compare that to what existing products can deliver. Is it enough? One of your specs was quiet. Was this quiet enough? The pumps you found didn’t fit in the box you had hoped for. Do you think you will be able to find small enough, powerful enough ones? This might be something good to spec before your decision
Execution
I didn’t look too closely at this, but it seemed to answer some questions!