SDG 3: Good Health and Well-being
SDG 11: Sustainable Cities and Communities
2. Project Details
Company or Institution
Environmental intelligence platform
General description of the AI solution
Meersens environmental intelligence platform is a digital solution based on AI enabling Meersens customers/partners to leverage the power of environmental data in order to improve populations health they are responsible for. Goal of the platform is to help deeply understand health outcomes caused by environmental hazards (pollutants) to deploy predictive and personalized prevention policies, given populations specificities and profiles.
The platform is based on environmental modeling, machine learning and explainable AI to render the most concise, relevant and understandable insights about environmental impacts, with a precision going to the street level and always in a predictive approach. Once an impact has been assessed, the platform recommends appropriate behaviors and prevention policies, in order to respect the 5P medical approach (preventive, predictive, personalized, participative, pertinent) it is based on.
The goal is to improve human health thanks to environmental and human processed data. Making the invisible visible is the first step to contribute to a healthier world.
Excellence and Scientific Quality: Please detail the improvements made by the nominee or the nominees’ team or yourself if your applying for the award, and why they have been a success.
Meersens is improving populations health through environmental and human data processing. The company has built an environmental intelligence platform that allows customers and partners to assess environmental hazards in order to preserve health of the population(s) they are responsible for. This platform addresses the exposome concept, and covers the whole cycle of environmental prevention, from risk assessment, to prevention policies recommendation. AI is at the core of this product on two particular fields:
• Environmental modeling: Meersens has built proprietary models in order to assess pollution (atmospheric pollution, noise, pollens) anywhere in the world in near real-time, historic and forecast. These models are coupling Bayesian methods, timeseries neural prediction and geographic context analysis to assess pollutants levels anywhere in the world, with a resolution up to 10m x 10m. Models are trained with more than 20 000 field sensors data, and constantly benchmarked to evaluate accuracy. For instance, average accuracy on air quality modeling (regarding EPA AQI classes prediction) is now 92%. TRL is now 9 and a specific algorithm (least exposed path) has been patented.
• Health trajectories prediction: Meersens is building models to predict health trajectories from exposure profiles, on three major health domains: respiratory diseases, cardio-metabolic troubles and mental health. Meersens has established a scientific collaboration with the LIRIS in France for machine learning, and with ISGlobal in Spain, for epidemiology. Goal of these models is to compute a risk score for the regarded health categories based on populations exposome and exposure profiles. Models are constructed from the data gathered by the HELIX and ATHLETE European projects (more than 32K individuals) thanks to explainable AI and will be transferred for the generic population with transfer learning. TRL for these models is now 4, as an international thesis is now ongoing between Meersens, LIRIS and ISGlobal on this subject.
Scaling of impact to SDGs: Please detail how many citizens/communities and/or researchers/businesses this has had or can have a positive impact on, including particular groups where applicable and to what extent.
Improving populations health through a better understanding of environmental effects is a key stake of our modern societies, insofar as these specific effects are tightly linked to the development of these latters. In particular, recent studies and work by the W.H.O have shown that more and more deaths are imputable to the environment and the conditions we live in, with nearly 25% of all estimated global deaths and 85% of chronic diseases being linked to the environment.
Having said this, Meersens solution clearly replies to the SDGs n°3 – Good health and well-being and n°11 – Sustainable cities and communities, with an innovative environmental approach.
In order to track Meersens solution outcomes, the platform includes various surveys and perception gathering processes, aiming to quantify the felt gain in terms of populations well-being. In addition, the platform is able to gather global indicators, such as illnesses prevalence indicators, showing on the mid-term/long-term the benefits of having deployed the Meersens solution and appropriate prevention policies according to area, environmental and populations specificities.
In order to deploy and commercialize its platform, Meersens has been focusing on the solution scalability and easiness of use. This results in a technical solution easy to setup, with no friction linked to the geographic area of deployment: assess environment, segment populations following exposure profiles, then observe model predictions and implement according and relevant prevention policies. The solution is now available in more than 180 countries in the world, with a setup timeframe of approx. one week to establish populations profiles, making it a wide and global solution at the UNESCO level.
Trying to improve human health through a better understanding of environment and populations specificities thanks to AI is definitely an innovative and breakthrough approach, that will contribute to the technical excellence of SDG ecosystem.
Scaling of AI solution: Please detail what proof of concept or implementations can you show now in terms of its efficacy and how the solution can be scaled to provide a global impact ad how realistic that scaling is.
Meersens already have evidence of impacts with first commercial deployments across private companies, cities & communities and even through scientific and health projects in Europe. In particular, the environmental intelligence platform is being used in two scientific collaborations to evaluate environmental determinants of specific health conditions: one with French INSERM regarding mental health, and the other with Luxemboug Institute of Health regarding long COVID symptoms. Our current slowdowns in terms of scaling up are mostly linked to the innovative nature of considering environmental data in daily activities. Indeed, for now these data are mostly being regarded as comfort or nice-to-have data, and deciders only begin to realize that it can have a huge impact in their activities, being the management of a community of workers or the administration of a town.
When targeting communities, local authorities or governments, we often notice that these communities are only beginning to use AI to address the common and recurrent challenges they face. Allowing them to use such technology in an easy and relevant way through the Meersens platform is a unique way to match sustainable development with technology, to understand these ones are better working together rather than being opposite.
Our technology is based on open-source components and our development teams actively contribute to this open knowledge: we are strong supporters of the open-source world. Regarding GDPR and data privacy, the platform is at the state of art of current techniques and practices, and any sensible data if fully protected at every layer of the infrastructure.
Ethical aspect: Please detail the way the solution addresses any of the main ethical aspects, including trustworthiness, bias, gender issues, etc.
The solution is fully ethical as it regards populations in their entireness, with no assumptions or distinctions being made regarding the gender, the race or the origin of these populations. The solution can be deployed anywhere without having to take into account cleaving characteristics, as our differences only result in different exposure profiles regarding environmental pollutants, which are being processed by the platform to compute health outcomes and suggest appropriate prevention policies.
The Meersens platform relies on the medical approach 5P (preventive, personalized, predictive, pertinent, participative) in order to setup virtuous circles for involved stakeholders thanks to personalized and predictive prevention.
The solution is fully lawful, ethical and robust.
Meersens improves Human health through environmental data processing, no matter where you come from and where you live.