Between 23rd-24th of April, MAMK organized Hackathon that continued the data seminar about combining AI and wellbeing on 22nd.

 

The challenges were

1. Medical imaging

2. Personal health and well-being

3. Creating value with professional medical data

4. Own track
At this Hackathon, all of the teams had project to solve real problems. The winning teams had most focus on developing tools for the challenges and had working prototypes with nice graphical designs.

 

1^st team:

AI.MD (Artificial Intelligence for Medical Data)

 

Developed tool to teach IBM Watson to recognize x-ray images (e.g for knee detection) by using Gamification.

 

2^nd team:

Coding Bad

Developed Health-assistant bot for Telegram, that you could communicate with text or speech and send image for the bot and it used Watson to detect you age/gender etc.

 

3^rd team:

Team Schrödinger

Developed way to detect if x-ray image quality is good or bad.

 

Other interesting team Fear Catchers were detecting emotions of user from multiple sources at same time by using many of the IBM APIs (Face recognition, Tone analyzer…etc) and then use that information to calm down the person (for example during presentation). Many teams also were trying to solve problems with obesity and diabetes (food diary, food imaging and mobile controlled wireless insulin injector). There was also working prototype Kettlebell that spoke (using speech synthesizer) to you how to use correct technique during workout.

 

 

IBM Bluemix API use example:

Tried one the APIs to recognize images. There is simple curl commands you can use to access the data (with your own APIkey)

curl “https://gateway-a.watsonplatform.net/calls/url/URLGetRankedImageKeywords?url=http://tethys.lib.helsinki.fi/pienpainate/binding/342795/thumbnail/1&outputMode=json&apikey={MyAPIkey}”

gave answer:

{

    “status”: “OK”,

     “usage”: “By accessing AlchemyAPI or using information generated by AlchemyAPI, you are agreeing to be bound by the AlchemyAPI Terms of Use: http://www.alchemyapi.com/company/terms.html”,

    “url”: “http://tethys.lib.helsinki.fi/pienpainate/binding/342795/thumbnail/1“,

    “totalTransactions”: “4”,

    “imageKeywords”: [

        {

             “text”: “person”,

             “score”: “0.960834”

        }

   ]

}

If you got interested, try the IBM APIs and register for free trial at here