Telehealth and remote vital signs monitoring is a booming sector
Developing a project whose core objective is to consume fitness tracking app (eg; Google Fit) data in order to allow monitoring of vital signs by

• user
• by their authorised friends and family
• by telehealth providers performing mass monitoring of signed-up customers, ready to arrange emergency response or callback as required

Metrics would be evaluated against the personalised AI model of user's historical metrics, in order to detect genuine anomalies for notification to authorised entities.

Open source alternative to premium monitoring services
Telehealth and remote monitoring is already providing a lower cost alternative to round the clock home care or hospital stays. This service aims to drive down the cost further with an open source project that leverages user's own mobile device as the source of metrics, for use where healthcare provider deems it acceptable.

Technical overview
Please check out the readme for overview and flowchart.

It will work by building a individual AI model for that user and then monitor user's ongoing health metrics by evaluating data consumed from say, Google fit API, with the model. Any anomalies are alerted on a UI available to user, along with notifications through other channels.

Stack is currently AWS ECS,
Sagemaker, Kafka, Flutter maybe for UI or Quick sight initially.

Getting involved
If anyone wishes to collaborate, add feedback, the Github project is at

https://github.com/jackpa99/iot_health_to_ai_aws