Kelsey M.

A main way academic economists build networks and share and get feedback on their work is by giving seminars at other departments. Departmental choices about whom to invite to give seminars can therefore shape individuals’ careers and may contribute to or help reduce existing disparities in the profession. I supported building this dashboard, which provides information about who gives invited talks, in aggregate, by department, and for each seminar series within a department.

The Student Success Tool is a product funded by Google.org and meant to scale student risk models to hundreds of institutions. I advised on and implemented scalable architecture, code, algorithms, performance metrics, and bias evaluation outputs such that data scientists can deliver unique insights and robust models to partner schools in a low-touch way.

The CMS Artificial Intelligence (AI) Health Outcomes Challenge was an opportunity for innovators to demonstrate how AI tools – such as deep learning and neural networks – can be used to accelerate development of AI solutions for predicting patient health outcomes for Medicare beneficiaries for potential use in CMS Innovation Center innovative payment and service delivery models.

I led a team of 5 data scientists, in collaboration with researchers and UX designers, to build a robust modeling and evaluation pipeline for submission. We predicted unplanned hospitalizations for Medicare beneficiaries within 30 days of a hospital visit, as well as 12-month mortality, based on a large data set of Medicare administrative claims data. We were also required to address implicit algorithmic biases that impact health disparities in our submissions. Our team was selected as a finalist.

I contributed a machine learning measurement, sensitivity at low alert rates, to an open source automated machine learning package.

The AHRQ Visualization of Social Determinants of Health Challenge invited participants to develop new online tools to present and encourage use of free, publicly available social determinants of health data to better understand and predict communities’ unmet healthcare needs.

I led a team of data scientists and researchers to build a submission to this challenge, which ended up winning the grand prize!