|
Activity Number:
|
91
- High Dimensional Data, Causal Inference, Biostats Education, and More
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 9, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
ENAR
|
|
Abstract #319052
|
|
|
Title:
|
Prediction of Relapse in Chronic Disease Using Fragmented Medical Records
|
|
Author(s):
|
Jiasheng Shi* and Jing Huang
|
|
Companies:
|
University of Pennsylvania Perelman School of Medicine and University of Pennsylvania Perelman School of Medicine
|
|
Keywords:
|
Markov Chain;
Grid;
Clustering;
Fragmented Functional Data;
Markov Modulated Process;
Gaussian Process
|
|
Abstract:
|
Most patients with chronic illnesses experience symptoms on and off throughout their life. Understanding differential remission and relapse risk profiles and predicting risk of relapses for individual patient are critical for optimizing disease management and improving long-term outcomes. However, medical records collected during clinical practice provide fragmented data with episodes of disease activities mostly observed after the onset of relapse. Disease activity during remission is often unobserved. In this study, we propose a method to improve the prediction of relapse using fragmented data from medical records. The method is applied to pediatric Crohn’s disease to better understand the remission and relapse cycle.
|
Authors who are presenting talks have a * after their name.
