Center for Clinical Quality & Outcomes Discovery and Evaluation (C-QODE)
DEMOCRATIZING DATA ACCESS FOR CLINICAL RESEARCHERS TO IMPROVE HEALTHCARE QUALITY AND DELIVERY
WHAT IS THE CQODE?
We aim to facilitate a pathway for young investigators to pursue data-centric research that improves healthcare quality and delivery. We accomplish this by democratizing data access for clinical researchers. We provide a platform for proceduralists to share resources to achieve economies of scale and facilitate new research projects that increase the velocity of evaluation, insights and discovery. Our mission is to develop and maintain collaborative multi-disciplinary and inter-departmental research teams that help scalable and deployable solutions to confront the healthcare system's most vexing problems.
The overarching aim of this center is to develop and implement a research training curriculum for junior faculty and academic trainees.
To fully realize the benefits of big data, we will combine medical coded data with text, interprofessional, imaging, and molecular data. This data infrastructure will be leveraged to support precision medicine, biomedical AI, and Implementation Science with the goal of improving healthcare delivery, reducing variation and cost of care, and ultimately saving lives.
The center’s focus is to translate new data into clinical and health decision making with a focus on:
Delivery of recommended timely care to the right patient
Population-level outcomes research that translates real health benefits to society
Providing infrastructure to the emergence of health services research encompassing surgical outcomes research, quality improvement, and implementation science.
Facilitating study design and initiation, grant and manuscript writing, and submission.
Democratizing data access for the UMN procedural researcher. This includes:
Following IRB approval no-cost access to pre-processed M Health Fairview EHR data in a HIPAA compliant environment for all patients Access to public use files and publically available datasets (MEDPAR, HCUP SID, HCUP NIS, NTDB, NSQIP, TQIP, SEER and others) in a HIPAA compliant environment that includes high-performance computing for faculty researchers including HIPAA compliant GPU support
C-QODE will facilitate federated data collaborations (EHR data, notes, images) with other institutions across the U.S.
USE CASE: "AMBISCRIBE" PARTNERSHIP
Intelligent Decision Support / EHR Population Project aka “Ambiscribe” Partnership of Dept. CSCI and Med School.
- Develop AI algorithm that can predict patient diagnosis based on historic data for ED chest pain admission
- Comprehensive series of simulated ED encounters and train AI algorithms to predict
- A prospective study with Surgery Clinical Trials Office to consent ED patients and providers using for the ambient recording of M Health Fairview ED encounters.
USE CASE: ROBOTIC SURGERY
Project Lead: Elliot Arsoniadis MD, PhD
The development of a Database of High-Quality Robotic Surgery Video and Kinematic Performance Data to mimic the surgical field, automate tasks, and train a surgical robot to complete a live synthetic test environment for surgical training certification.