Eunai - Space Station Contagion & Health

EUNAI - Space Station Contagion & Health

print("Hello, new_World!")

Inspiration

Humanity has moved entirely to life in space stations. With this new reality in mind, our team brainstormed what hurdles this new frontier would entail: the logistics of a space colony, asteroid showers, radiation, adapting the human body, missing nature, space pirates... the dangers are endless. But we stood out to us most of all was the terror of...

... a common cold?

What would be a minor inconvenience back on Earth may actually be disastrous in a space station. The cramped, permanently enclosed quarters on top of the human body lacking nourishment from the sun and nature, meaning weaker immune systems without additional supplements, would bring disaster to the highly fragile logistics of running a space station.

What it does

EUNAI is a smart prediction system aimed to prevent the spread of disease in a space colony. Prevention is better than cure, after all. EUNAI takes into account your biology, such as your age, average exposure to artificial sunlight and sleep cycles among other factors, to estimate how likely you are to become ill. It will alert you if you have visited a sector recently where another person is reported sick, giving you a prediction of how likely you are to have caught it and then take the necessary measures. Additionally, using EUNAI's webapp will allow you to update your biological details and view your health status at any time, on any device. EUNAI also pairs with popular messaging services if you do not wish to install an app, allowing you to directly access your health status, update your data, report sickness or manually upload your location if you do not wish to be tracked! This allows those with disabilities to access accommodations they are already familiar with within their preferred messaging service.

How we built it

The system uses linear regression machine learning to create a custom predictive model. This was done using python, allowing us to handle a custom dataset used to train the model. A custom dataset was required due to factors that are unique to life in a space station, such as artificial sunlight.

Python was further used to access Telegram's bot services, allowing us to create a functional telegram bot for users to interact with, which links with our predictive model.

Wix was used to create a dynamic webapp that adjusts to different screen orientations and sizes, as another method for users to interact with our model, giving them the benefit of a choice.

Challenges we ran into

Due to the time constraints of the hackathon, it was a challenge implementing all the features we wanted, especially since it was our first time creating and training a custom machine learning algorithm and model. With that being our biggest hurdle, the ease-of-use of python allowed us to code and test quick, and Wix allowed us to build an interactable webapp as a proof-of-concept in time.

We also ran into issues working together as a team, since we had our personal commitments to attend to during the hackathon, disrupting our workflow. Despite this, we managed to communicate well and utilize Git to work together without being in the same room.

Accomplishments that we're proud of

This was the first time Ramses, a member of our group, used Python, so having learned a new programming language enough to create a functioning telegram bot with many working commands within the time period of the hackathon is something we're proud of. For Saksham, it was his first time building a website that was both aesthetically pleasing and highly interactable, as well as dynamic across screen sizes, which is an amazing accomplishment. With Anuk, he has never approached handling databases or creating a machine learning algorithm before, so having built a custom model that still has a high degree of accuracy is another point of pride.

What we learned

As a group, on top of the hard skills we learned with Python, Wix, database handling using APIs, etc... we learned alot about efficient communication and working together as a team - delegating tasks, resolving conflicts, playing to our strengths.

What's next for EUNAI

The custom model has been trained on factors that affect the common flu, but given more time we hope to expand this to a range of contagious illnesses. Linking to verified medical professionals within the app itself, so that a user can get professional advice or book a diagnosis appointment, would be another great addition to the system. Further, we realize that terms such as 'obese' may cause discomfort for users, and the nature of such a service may cause anxiety, so we hope to include accommodations for mental health as well.

GROUP Anuk Ahangamgoda 62872734 Saksham Joshi 60442787 Ramses Remus Basrie 24252314