Our little brother he is a Paramedic and Medicine student he identified an oportunity to improve the full process of Medical asistence, by keeping the patience with a known situation that require monitoring to act fast, and keep the "Gold hour" on track to have the chance to save more lives.


Accidents can happen at anytime and anywhere, on the street, at home, and in workplaces. The consequences for some type of accident can be multiple for both the sufferer and the family. Rapid care after the incident ensures more chances of surviving and / or avoiding permanent sequelae. This is a system that does not intend to change the protocols of reception and management of medical emergencies but to improve them and make them more efficient with shorter response times, automatic identification, etc.

What it does

The system keeps receiving information from the health bracelet of the user. It takes the information and compare against the limits that each person and the medicine dictate as control values. If the combination of values detect an issue, it makes contact with user and decide to send the closest ambulance automatically or join the call to an agent that will takes the desicion.

How we built it

The user that recognize need to have medical aistance unexpectly gets the HealtLine Iot Bracelet and gets subcribed by the MobileHealthL application application. We can see the info on "WebHealthL" now. This is a Javacript, Jquery, and Html application. Using "AWS Cognito Pool" as authorization.

The bracelet keep sending data about the vital signs to AWS Iot. In this case we are using a simulator build on NodeJS running in a EC2 instance that is sending the data to the cloud system. The vital signs send are: { id: , clientid: "1002", heartrate: hrate, temperature: temp, location :"25.583717,-100.2572317", datetime: new Date( } using topic:healtvitalvalues

AWS IoT receive the information constantly and has three rules, the first one gets the data under topic "" and send it, separate and insert into the DynamoDB table "dinamoIOT". second rule keeps sending data to an Amazon Kinesis Stream "iotkinessis2", this stream keeps been consumed on real time "WebHealthL", the third one is invoke the Lambda Function "function_iotv1" when it detect the heartrate is above the minimun Level, on this case 70. Is expected on real production the levels be variables, personalized and all the other vital signs be monitored. The Lambda function writen on NodeJS is in charge to perform a call to the monitored user, Using Twilio to perform the call and ask the user to click on key 5, the system detects the response, and save the info into a log to learn. If the user does not answer or click the correct key, in real production would be expected the system do a retry but now calling also to the second contact provided, both does not answer correctly or one of them click the fast key.

The system on this case handled by the Lambda "TwilioGatherLambda" fucntion will detect this and start the next phase, on KeyFast key, the system would automatically send the most closest Ambulance, because all the float will be connected too, will arrive to Monitored User quickly as the system has the to the gps location.

On the situation of non or incorrect answer, the Human Medical Agent on "WebHealthL" will be notified by "AWS SQS" and also will receive a call, that will automatically be join to the Monitored User, if answer, Medical Agent will evaluate the situation and click "Send Ambulance" on the application that will use same logic that KeyFast. Everything keeps recorded on DynamoDB "trackinghealthline".

Challenges we ran into

New on Tools , every step was a challenge to dicover at the end that there was not muchdifficult

Accomplishments that we're proud of

Really proud of sumit it. Wanted to make it better and better but still long way to walk on it.

What we learned

We have been learning a ton. We understan that "CloudWatch" is our friend.

What's next for HealthLine

A side of the technicals changes and improvements absolutelly required, HealthLine also will be complemented with a Module of Voice and Text interaction using "Amazon Alexa" for the paramedics that arrive to the scene to be able to capture the current values and situations and have them delivered directly to the Hospital where Monitored user is send. As this is another point where valuable time and Information is wasted.

Built With

Share this project: