Skip to Content

Team Giraffe

2020
Team Members:
  • Jiayuan Kong
  • Ran Sui
  • Yinghe Sun
  • Tony Wei
  • Evan Yu
Advisors:
  • Lee Andrew Goeddel, MD
  • Raimond Winslow, PhD
  • Joseph Greenstein, PhD
  • Han Kim
  • Hieu Nguyen

Abstract:

Acute kidney injury (AKI) occurs when decreased renal perfusion or toxins cause glomerular or renal tubular damage. While often somewhat reversible renal injury can progress to renal failure requiring mechanical replacement of kidney function. AKI is associated with longer hospital stays and costs, higher incidence of chronic kidney disease, and increased hospitality mortality rates by up to ten fold. More than 10 million people worldwide suffer from AKI each year, of which between 30 to 40% occurs post-surgery.

We have data from a retrospective cohort of 10,123 surgery patients cared for at the Johns Hopkins Hospital from the aQI (anesthesia quality improvement) dataset. Features from aQI include demographic data such as age and race, time series data such as blood pressure and heart rate, and other parameters such as medications taken, lab results, and comorbidities.

We propose to create a machine learning-based platform to give real-time risk assessment of AKI, which would cut down on human resources costs, give clinicians the opportunity for early intervention, improve patient outcomes, and decrease health care costs.

Read the Johns Hopkins University privacy statement here.

Accept