Progression of chronic kidney disease in African Americans with type 2 diabetes mellitus using topology learning in electronic medical records

Abstract

Author contributions: JS, BIF, DWB, and MCYN initiated the study; JS and LW designed and performed the data analysis; XZ, LSH and JX prepared and preprocessed the data; FCH and SHC supervised the statistical analysis; JS and LW prepared the manuscript; BIF and DWB revised the manuscript.All rights reserved. No reuse allowed without permission.was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which . http://dx.doi.org/10.1101/361956 doi: bioRxiv preprint first posted online Jul. 4, 2018; 2 Abstract Background. Chronic kidney disease (CKD) is a common, complex, and heterogeneous disease impacting aging populations. Determining the landscape of disease progression trajectories from midlife to senior age in a "real-world" context allows us to better understand the progression of CKD, the heterogeneity of progression patterns among the risk population, and the interactions with other clinical conditions. Genetics also plays an important role. In previous work, we and others have demonstrated that African Americans with high-risk APOL1 genotypes are more likely to develop CKD, tend to develop CKD earlier, and the disease progresses faster. Diabetes, which is more common in African Americans, also significantly increases risk for CKD.

DEPOT: graph learning delineates the roles of cancers in the progression trajectories of chronic kidney disease using electronic medical records

“…With the learned representation, we then use the reversed graph embedding to learn disease progression trajectories from EHR data. As reversed graph embedding approach [6,25,26] assumes that the longitudinal data points are fragmental and sampled from the underlying trajectories (mathematically known as the principal graph underlying the observed fragmental data points), thus, it is intrinsically suitable for learning trajectories from temporally fragmental clinical observations. By extending our previous work [26] to systematically outline the progression landscape of CKD, our DEPORT approach casts new light on EHR-based clinical research.…”

“…As reversed graph embedding approach [6,25,26] assumes that the longitudinal data points are fragmental and sampled from the underlying trajectories (mathematically known as the principal graph underlying the observed fragmental data points), thus, it is intrinsically suitable for learning trajectories from temporally fragmental clinical observations. By extending our previous work [26] to systematically outline the progression landscape of CKD, our DEPORT approach casts new light on EHR-based clinical research. The DEPORT approach addresses the two major limitations of EHR data analysis and fully unleashes the power of the highly incomplete and informatively censored EHR data with heterogeneous features.…”