Abstract: TH-PO012
A Data-Driven Digital Health Approach to Improve CRRT Delivery in the Intensive Care Unit (ICU)
Session Information
- Augmented Intelligence for Prediction and Image Analysis
October 24, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Authors
- Raposo, Ruben, Dialytix Technologies Inc., Birmingham, Alabama, United States
- Neyra, Javier A., The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Tolwani, Ashita J., The University of Alabama at Birmingham, Birmingham, Alabama, United States
Background
Acute kidney injury (AKI) is a frequent complication in the intensive care unit (ICU). About 15% of critically ill patients with AKI require renal replacement therapy, commonly in the form of continuous renal replacement therapy (CRRT). The aims is to develop a data pipeline to feed scalable dashboards of patient- and CRRT-relevant data to monitor key performance indicators and patient outcomes.
Methods
We utilize EHR and CRRT device data from the University of Alabama at Birmingham from January 2022 to February 2024. The data from each of the 42 CRRT machines (2022 to 2024) operating UAB. We extracted the CRRT device data every month via a manual process and were then uploaded to a secure storage location for data processing. We created an ETL (extract, process, and loading) data process using Python code to clean and normalize data. We then connected the Python data pipeline to feed the visualization layer via the Tableau cloud server. We designed interactive dashboards with Tableau that show the key performance indicators for various stakeholders that want to monitor and track on a weekly or monthly basis.
Results
We developed clinical dashboards in Tableau that incorporate data visualization with KPIs, filters, and AI-powered data analytics techniques to support providers in viewing and exploring EHR data for clinicians to inform decisions and improve patient care, thereby identifying previously unnoticed patterns in EHR data to support clinical decision-making.
Conclusion
With the collection of CRRT machine data and the creation of a data processing pipeline including EHR data, we can provide our clinical, administrators, and medical researchers further data insights on the performance of the CRRT treatment and enhance the monitoring capabilities so that we can continue to understand where possible gaps are and find ways to improve on them.
Funding
- NIDDK Support – Dialytix Technologies Inc.