On 12 July 2017, the UNICEF Innovation Fund announces another set of investments in open source technology solutions – Kimetrica is among one of three new portfolio companies

The UNICEF Innovation Fund invests in technology start-ups from developing markets that are working on open source solutions to improve children’s lives. The Innovation Fund applies a venture capital approach to source solutions that can impact the lives of the most vulnerable children. These solutions are clustered around $100billion industries in frontier technology spaces.

Check out www.unicefinnovationfund.org for more information – including real-time data – on each investment.

Kimetrica using facial recognition technology to detect malnutrition in children

How would you describe your solution?

We at Kimetrica are developing the MERON (Method for Extremely Rapid Observation of Nutritional Status) app, which uses facial recognition technology to detect malnutrition in children (aged 0-5) during humanitarian emergencies.

MERON can be downloaded to any tablet or smartphone with a camera. All MERON needs is someone who can take a picture. Once the app snaps a child’s picture, that image can help us determine his or her health status. MERON then uses an algorithm to analyze facial curvature and assesses other nontraditional markers to estimate body mass index. For child safety, the actual image is not stored, merely key points of the face that will be used to create the facial map. This information can assist us in identifying those children who need nutrition support and in getting it to them in a timely manner.

MERON is therefore intended to be a scalable alternative to the Mid-Upper Arm Circumference (MUAC) rapid assessment method, which requires training and supervision to reduce errors in application.

What is unique about your solution and what is the competitive landscape like?

Our solution will allow for faster data collection than current methods do – and for children, MERON is a less intrusive way of measuring their nutritional status. MERON has the ability to map and link the curvature of the face to Body Mass Index.

Why does open-source technology make your solution better?

MERON will be provided through an open API. Through open sourcing, any organization in the humanitarian assistance sector can access, share and easily integrate MERON data with other kinds of data collection, thereby improving systems to better detect malnutrition. MERON is agnostic; it can work with other systems, tools or products with ease.

How did you come up with your solution and what inspired you to form your company?

It is fitting that a child would inspire an app to help other children. Meron is named after one of our founder’s daughters. In 2016, we were at dinner and discussing the challenges of using MUAC in an emergency environment, when Meron who was 14 at the time asked: If you can use an arm then why can’t you just snap a picture of a face, like Snapchat, and calculate the kid’s health that way? It was a lightbulb moment and although we were skeptical at first, we read the literature and saw that she was onto something.

How did your team come together? What is your team’s MO and drive towards the problem you’re trying to solve?

We are a small part of a big team at a research firm called KIMETRICA. Our founders, Ben Watkins and Rob Rose, are former multilateral organization and humanitarian assistance employees who saw a need for evidence-based research in the area of humanitarian emergencies. What motivates us is a desire to figure out more efficient ways to get aid to the people who need it most.

What do you plan on doing with UNICEF’s Venture Fund investment and how will you use that to leverage raising follow-on investment?

We are using this investment to get to the next phase of development, which is collecting data to test MERON, or “train the model,” on a large sample of our target population. We will work in parallel with UNICEF nutritional teams–who are using the gold standard methodology, “The Weight-for-Height Z Score”–to collect data on three physiognomies and compare findings. Ultimately, we aim to conduct a real-world test and submit those findings to a peer-reviewed journal.

print