Saral

COVID-19 has brought attention to the significance of public health and safety. While our daily lives have been disrupted for over a year, we've learned to adapt healthy practices (as well as a slew of new words!) and accommodate new routines. When it comes to COVID-19, the elderly are particularly prone to severe sicknesses.

According to research, adults 60 and older, particularly those with previous medical problems such as heart disease, lung illness, diabetes, or cancer, are more likely than other age groups to develop severe — even fatal — coronavirus infection. There are bound to be multiple projects based on making hospitals and other such gathering areas healthier, but our project aims to improve the immediate environment of the elderly, regardless of where they are.

Our solution, "SARAL." is a wearable device fitted on a belt with its companion app; that can give alerts on fall detection,measure heart rate and SpO2 levels, and send them to servers for medical practitioners to investigate. It also gives off signals when the user is stagnant and can be configured to give rich formatted alarms for medicines and other activities. We can also use it to load medical reports ( by family members and registered medical professionals) via a protected QR code. With a simple-to-use UI on both the device and application sides, we believe SARAL would stay true to its name and make lives easier for our elderly friends.

AWS Structure from IoT Side

AWS Structure from the App SIde

Features

QR Code Report Scan:

QR code is stored on the M5 Stack that points to the user's reports stored on s3 bucket which shows the latest report uploaded by the doctor

Fall Detection:

We use the internal IMU’s accelerometer to detect falls.

As soon as the accelerometer detects a value less than 0.4 it sends a true value to the IoT Core

An IoT rule is configured which sends the data to AWS SNS which then sends an Email to the email-ids subscribed to that topic in SNS

In the SNS-based notification system, the user can also configure an SMS instead of an email.

IoT Rules that form the guidelines of the on-event actions.

Demonstration of the Simple Notification Services (AWS)

Final Message that comes on the registered email -We can change the recipient(s)d according to the user's wish.

Heart Rate:

We use the Heart Sensor details such as Heart Rate and Spo2 data from the MAX30100 sensor and Upload it to the IoT Core.

From the IoT Core, we have attached a Lambda Function that converts it into a JSON Object named HealthParams and adds it to a column inside a table named Saral-aws-data where the Device Id is the Primary Key.

Unfortunately, our Max30100 wasn't working so we have commented out the max30100 code but we have tested the AWS side from the Test MQTT Client and are sure that it works.

IoT Rules Pages

Lambda Function

DynamoDB table where the Object Health Params is appended

Stagnancy:

No change in movement triggers the vibration motors to spin and turn the side LEDs to red. This is a local feature that doesn’t require IoT to function.

API Endpoints: ( {operation: operation, payload: payload} )

register: takes in the email and password and initializes the user in the database

login: takes in email and password and returns a JWT token that can be used to access the remaining API

uploadReport: creates a signed s3 upload endpoint which is then used to upload a PDF onto the database

me: this endpoint is used to display the user information on the mobile applicationdisplay the user information on the mobile application

Read QR code: this endpoint is triggered when someone(say a doctor) reads the QR code with their phone. It sends a 301 redirect to the last report in the user’s database

Mobile App:We built a mobile appt on Flutter. The first steps were to develop a wireframe and then creating a clean and straightforward UI on Figma, which was designed, keeping in mind the two driving factors of our project: ease of use and a rich feature base.

Potential Future Work:

Make the fall detection feature send a message + show a notification on the mobile device

• Make the fall detection feature send a message + show a notification on the mobile device
• Have an ML algorithm for Fall Detection

Demonstrations :