Medbay medical kit.
Main view. Summary of treatments and follow-up status.
Medicine details and progress of the treatment.
History of treatments.
Manager view. Read QRs and manage prescriptions. Sign them with Signaturit.
After attending this year’s MWC we realized that connected living can be quite a challenge for the elder. This is why we have focused the development of our hack on building what could be the next generation of home medical kits, a smart device that aims to ease the follow-up of medical treatments and improve the patients' experience. Say hi to Medbay, the AI-powered medical kit!
What it does
Medbay is an AI-powered medical assistant that has been developed to help patients, families and doctors by making the tracking of ongoing treatments easy. This medkit has been built using Amazon’s Alexa so it can understand the user’s symptoms via voice recognition and offer a non-prescription drug that is automatically delivered. Its companion app is meant to be used by doctors and caregivers to keep an eye on the progress of the different treatments, track the medicine intake and receive alerts in case the device is being used by someone else.
Medbay consists on the medical kit, which is the hardware, and a mobile app for doctors that’s linked to it. This way, they can interact with the system and monitor the patient’s drug use. The system is able to understand your symptoms and offer you a non-prescription drug based on them. In the tracking page of the app, the doctor can see information of the daily progress of the patient, assessing if he or she sticks to the treatment. A part from the database of prescriptions and the inventory of stored drugs, having the history of symptoms and drug consumption habits will help the doctor readjust the treatment and even find patterns of misuse. The app also allows for the integration of prescriptions. Doing it is as simple as filling a form or just scanning a QR code from the prescription. This QR contains information on the name of the drug, the dose and the schedule. Then, this is signed in a secure and legally-binding way thanks to Signaturit API.
How we built it
To build the Medbay we have used everything in our toolbox, from hammers and saws to stepper engines, Alexa’s API, Go Language for the programming, Progressive Webapp development, Sketch and Illustrator to design the whole UI.
Challenges we ran into
We ran into several challenges while developing and deploying the project. First of all, although we had reserved Amazon Echo a week before the event, they were gone by the time we got to the hardware depot. That implied that we had to configure Alexa (Voice and Skills) from scratch. After that, we had to figure out how to setup the engines, we were having some issues to make them work the way we wanted them to. Related to the design of Medbay itself, we encountered the kind of problems teams face when developing hardware: finding the right parts, figuring out the mechanisms and putting it all together when finished. To design the UI we used Sketch and, since it was our first time using this tool, taming it took us a bit of time.
Accomplishments that we are proud of
First of all, we are proud of submitting the project on time, developing a piece of hardware always adds a layer of complexity to the problem. Managing to setup Alexa without Amazon's Echo would be another thing we are very proud of. It was also the first hackathon where Ricard and Adrian had to get their hands dirty with code and we think it shows a progression of the team, being more cohesive hackathon after hackathon.
What I learned
Prototyping techniques, consolidated web design fundamentals.
Go, Alexa API, Python, Sketch, Illustrator, HTML, CSS, Scalingo, Signaturit API.
What's next for Medbay, the AI-powered medical kit
Pattern analysis of the medicine intake of each patient to help the doctors' workflow. Build a better prototype (we think it's pretty cool as it is). Add notifications to the APP ("Call me if I forget system"). Add a display to the medkit to improve usability and track of medicines stored. Biometric data acquisition and analysis through a wide variety of sensors (IR for temperature, galvanic skin response, heart rate sensors, blood sugar analysis and insulin dispensers).