Skip to navigation Skip to main content Skip to footer

Approved Research

Using EHR data to enhance the analysis of multi-modal data

Principal Investigator: Professor Nima Aghaeepour
Approved Research ID: 106206
Approval date: December 12th 2023

Lay summary

This research project is focused on using artificial intelligence and machine learning to better predict the risk of diseases like heart disease and diabetes. The program uses a type of machine learning called deep learning, which is a way of teaching algorithms to find patterns in large amounts of data.

We plan to do this in three main steps. First, we start by using deep learning to analyze standard health data from tests like blood tests. Then we incorporate clinical data, which includes information about a patient's overall health and history, to improve our initial analyses. Finally, we use additional clinical data to further increase the accuracy and power of our analyses. This research is designed for situations where there's a lot of clinical data, but not much of the other kind of data, like from blood tests or biological assays. This is a common problem in this type of research, and this new method could improve analyses in all of these studies.

We are also interested in understanding more about how diseases develop. By looking at both clinical and blood test data together, we hope to gain new insights that wouldn't be possible by looking at either type of data alone.

The predictive models and disease insights could have a large impact on public health. By improving our ability to predict disease risk, we could better identify people at risk and provide them with early treatment. This could help prevent diseases and improve health outcomes for many people. In the long run, this research could also help us understand more about how these diseases develop, which could lead to better treatments and maybe even cures in the future.