Namaste! A couple of weeks ago I wrote about Deep Learning in humanitarian work. Welthungerhilfe has the righteous goal to reduce world hunger to zero in the near future. Their project Child Growth Monitor is a very promising project that is definitely going to help. Yesterday, at our Würzburg School of AI Meetup, we talked about the project at length.
ZDI is the place to be.
We met at ZDI Mainfranken for the very first time. Their ideation lab is in the tower of the former barracks. It is usually the place for a huge variety of workshops. If you want to be creative, do a lot of brainstorming, fuel imagination and innovation, this is where you want to be if you are in Würzburg.
Since its opening in July the project team has spent quite some time at the ZDI. Working on the project. Deciding on core features. Sharing ideas about future work. Doing experiments. Keeping the project on track. ZDI in Würzburg is an excellent environment for such endeavors.
Markus Matiaschek is the man.
Markus Matiaschek is an overall nice guy. And he is the founder of Child Growth Monitor. It was my pleasure to introduce him to the stage at the meetup. Markus gave an in-depth introduction to the project. He provided an outline of the software architecture and a roadmap for future work. And he also talked a lot about lessons learned and challenges.
Truth be told. The project teaches everyone who has the chance to contribute a lot. About the current state-of-the art. About next year’s state of the art. And about the fact that the current times are awesome vor great and innovative projects. I am very happy that Markus is the lead behind one of those.
A use-case for Deep Learning?
Is Child Growth Monitor a use-case for Deep Learning? Let us think about that… The idea is simple. Nomen est omen. The name explains the purpose. If you manage to monitor the growth of a child properly, you can fight and prevent malnutrition. Unfortunately measuring and monitoring is a crucial issue. First of all, there is too many children in need. And second of all, the standard measurement process is very, very error-prone. Why not using an app with a decent camera to do the measurements in no time?
On the first glance, Child Growth Monitor is definitely a Deep Learning use-case-candidate. You got a lot of input-data. That is for sure. Images and point-clouds of many children. And you got a lot of output-data. Manual measurements like height and weight. All that is missing is a proper pattern detection algorithm. Classic Deep Learning.
If it really is a Deep Learning use-case, we do not know yet. But we have a very high confidence that it is. Why? Our experiments from the last couple of weeks definitely pointed in that direction. How will we find out for sure? Doing more experiments! The data is coming in high amounts on an every-day basis. Every experiment we are going to make is going to contribute.
Another good thing about Child Growth Monitor: The project is very open for collaboration! Join and star our GitHub-repository. Follow us on Twitter. And drop us a line at info(Replace this parenthesis with the @ sign)childgrowthmonitor.org, if you like. We are looking forward to hearing from you!
Stay in touch.
I hope you liked the article. Why not stay in touch? You will find me at LinkedIn, XING and Facebook. Please add me if you like and feel free to like, comment and share my humble contributions to the world of AI. Thank you!
A quick about me. I am a computer scientist with a love for art, music and yoga. I am a Artificial Intelligence expert with a focus on Deep Learning. As a freelancer I offer training, mentoring and prototyping. If you are interested in working with me, let me know. My email-address is email@example.com - I am looking forward to talking to you!