Kedir Turi

Vanderbilt University Medical Center

1 active project

Asthma Attack Prediction: Machine Learning Approach

Despite progress in therapeutic approaches, according to CDC about 50% of individuals with asthma continue to experience asthma attack (Asthma exacerbation) every year. Asthma attack is an acute episode of progressively worsening of asthma symptoms. Uncontrolled asthma (episodes of asthma…

Scientific Questions Being Studied

Despite progress in therapeutic approaches, according to CDC about 50% of individuals with asthma continue to experience asthma attack (Asthma exacerbation) every year. Asthma attack is an acute episode of progressively worsening of asthma symptoms. Uncontrolled asthma (episodes of asthma exacerbation) can have a significant impact on patients' quality of life, healthcare cost and burden, and time away from school and work. Predicting when exacerbation may occur with adequate intervention would reduce healthcare burden, save cost, and improve quality of life. However, efficient methods that can help identify personalized risk factors and make early predictions are lacking. The objective of this project is to use machine learning models to better predict the risk of asthma exacerbations and calculate saved healthcare cost. All of Us Research database with about 35,000 asthma diagnosed patients and about 10,000 asthma attack events provide an opportunity to build a prediction model.

Project Purpose(s)

  • Disease Focused Research (asthma)

Scientific Approaches

We will retrospectively build a cohort of asthma patients and follow them for asthma exacerbation events. Exacerbation events will be defined as those who needed 1) an oral glucocorticoid prescription for less than 28 days (glucocorticoid burst), 2) an emergency department visit, or 3) hospitalization. Predictive models will be built using a gradient-boosting-machines framework. We will attempt to predict by healthcare utilization type (emergency and hospitalization). Prediction model will be built for pediatric and adult age groups separately. For each outcome, 80% of the dataset serves to train the model and 20% to validate it.

Anticipated Findings

The out come of this project is asthma exacerbation prediction model that is trained and validated on individuals from diverse geography and communities in the US.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Registered Tier

Research Team

Owner:

  • Kedir Turi - Other, Vanderbilt University Medical Center
1 - 1 of 1
<
>
Request a Review of this Research Project

You can request that the All of Us Resource Access Board (RAB) review a research purpose description if you have concerns that this research project may stigmatize All of Us participants or violate the Data User Code of Conduct in some other way. To request a review, you must fill in a form, which you can access by selecting ‘request a review’ below.