Agex: Elderly care chatbot

Inspiration

Elderly care has always been an action point in the field of geriatrics. With 46 million people in the US in this age group and the need of them to stay at home results in some form of depression in around 10% of the population. This results in a need for all around clock care for the elderly population. As mentioned all around the clock care is needed for the elderly population. At home nurse assistance is costly and not all can afford it. But with smartphones being present with almost everyone but with the issue of no common point to meet all needs at any time of the day. Also none of these apps are specific to the elderly population. And they being not so tech-savvy there is a huge need for a platform which is easy to navigate. As mentioned there is a need for an empathetic but vigital virtual assistant which can just act as company or even provide the necessary help in times of need.

What it Does

Here the development of a platform for all on demand services for the elderly i.e A- X Webapp service : AgeX is shown. Through this tool the elderly can easily navigate through the different features like nursing assistance which eventually calls the required nurse/physician. Along with sending in a confirmation email and notification pop-up. As for other services which are geographical domain specific like on-demand food services can also be enabled through this feature. The highlight of the tool is a chatbot to provide companionship to the elderly whenever needed through to & fro empathetic conversational statements and questions; which is also connected to the previous feature by assisting features like contacting the physician/therapist or advice for nearby restaurants. There is also an SOS tool in need for help. Lastly there is a feature to call a companion whenever in need like just to have a fun conversation or assistance for a certain task is required, There is a database of such companions/ volunteers through which such a feature is run.

How we built it

The application was built as a responsive web application, compatible to be deployed as a PWA using container view for native applications on iOS and Android platforms, using the following tech stack: For BackEnd services : Python-Flask Deployed : Google Cloud Platform Templating engine : Jinja FrontEnd : HTML + CSS Google Maps API Chatbot Flow : Lex Google GeoLocation API Google Places API Google Cloud SQL Service Google GeoCoding API The app starts with a splash screen, and a login prompt. At this time, we decided to regulate our users on the platform and make it invite only, barring the registration page. Next, the user is greeted with 3 options: Get On - Demand Service (Nurse / Medicine / Care) Chat with our empathetic chat bot Get Companion for Social Events / Online The flow from here on depends on the option that the user selects. If they select the first option, the application accesses the current location of the user (After asking for permission) and offers on demand service, giving the option to schedule a service and giving the confirmation code along with mail notification. After choosing the second option, the user is taken to our empathetic chat bot, where they can chat, and understanding the context of the conversation, the chatbot gives them options to schedule events, find companions, get on-demand service, and understand if the user wants food or is not feeling well. The third option is our (in progress) chat with nearby like minded elderly people, to do which, they need to request to connect first, and upon approval, they can connect. The same list is used in close conjunction for other features as well. Throughout the application, we have given the user an option for SOS and using clutter free UI, an easy user experience.

Challenges faced

Cloud Deployment of SQL instance was something which was really difficult, especially testing the same on local system through the use of cloud proxy script. Integration of Google places API and creating it context-aware.

Accomplishments

A web app was created using various tools which would really be helpful for not only during the time of pandemic but also in day to day life.

What we learned

We learnt a lot of things during the course of this Hackathon. A few of them are:

  1. Progressive web app development
  2. Using Google Cloud Platform
  3. Building a chat-bot
  4. Integration of google’s API into our web app

We also developed our collaborative skills during the course.

What's next

Some of the things that we want to try in the near future:

  1. Generate continuous behavioral analysis reports with respect to responses by the patient. This will provide us with important insights into the overall well being of the user.

  2. Different profiles for user and concerned caretaker to allow improved tracking of responses and needs

  3. Voice enabled navigation

And more importantly take user feedback and work continuously towards improving the experience

Important: Logging Info Login with username: test@gmail.com and password: test

Built With

Share this project:

Updates