Welthungerhilfe, one of the largest private aid organizations in Germany, is always looking for innovative approaches that help to effectively combat hunger in the world. The organization is currently testing a mobile app in in India that accurately measures children using 3D imaging rather than manually weighing and measuring the children. The goal is to use the technology for the precise assessment and identification of malnourished children. Welthungerhilfe has been looking for and found a worldwide cloud platform for the development and operation of this app: the Microsoft cloud. This will enable developers to evaluate 3D scans with deep learning models from Azure’s AI services. In addition, Welthungerhilfe needs partners with in-depth analytical know-how, that’s where Munich Re kicks in.
In mid-November, the reinsurer, together with Social Impact Partners, Microsoft and Welthungerhilfe, called for its a Datathon under the motto: “Identifying and combating hunger risks more effectively with Artificial Intelligence”. Big data experts, analysts and developers were invited to develop innovative solutions around the AI algorithm in Munich Re’s LAB. Following our goal ‘Building a safer future together’, ITHM has sent a team of AI and Big Data specialists to the ‘race’. In 48 hours interesting solutions came about. Our team for example was able to improve the existing neural network so that the size of the child could be accurately measured to +/- cm 5. In total, over 40 Data Scientists and Engineers participated in the event. All results were made available to Welthungerhilfe in order to advance the application to market maturity.
Thank you for your engagement!
Niko Tsanakas, Anna Schröder, Tim Matzeck (Inverso GmbH), Sandeep Sinha and Manuel Holzhauer (InusrTech Hub Munich)