Better and more representative labeled data allows for the potential for machine learning to improve access to healthcare and outcomes worldwide.
COVID-19 has laid bare disparities worldwide in access to quality preventative services, diagnostics, and care, as well as structural factors that contribute to vulnerability to illness. Yet across the variety of ML-enabled solutions being proposed to address the COVID-19 pandemic and beyond, many applications rely on large labeled datasets that are built in the Global North, or only pertain to segments of the global population.
The Steering Committee of Lacuna Fund is considering how it could support the expansion of existing datasets related to COVID-19 and respiratory illness response to make them more representative of and therefore more useful to all people affected by this virus. In particular, we want to make sure that any additional data that is collected or labeled contributes to solving problems identified by healthcare professionals, local-decision makers, and other frontline workers.
In addition to health impacts, Lacuna Fund recognizes the dire food security implications of the pandemic and will strive to align its funding for labeled datasets in the agriculture and language domains to create both short- and long-term impact that helps alleviate the impact of the crisis.
See information about open application processes here.