Using Machine Learning to Predict Acute Kidney Injury in Patients Following Cardiac Surgery
Overview
- Phase
- Not Applicable
- Intervention
- Not specified
- Conditions
- Machine Learning
- Sponsor
- Yunlong Fan
- Enrollment
- 2108
- Locations
- 1
- Primary Endpoint
- acute kidney injury
- Status
- Completed
- Last Updated
- 4 years ago
Overview
Brief Summary
Cardiac surgery-associated acute kidney injury (CSA-AKI) is a major complication which may result in adverse impact on short- and long-term mortality. The investigatorshere developed several prediction models based on machine learning technique to allow early identification of patients who at the high risk of unfavorable kidney outcomes.
The retrospective study comprised 2108 consecutive patients who underwent cardiac surgery from January 2017 to December 2020.
Investigators
Yunlong Fan
Clinical Professor
Chinese PLA General Hospital
Eligibility Criteria
Inclusion Criteria
- •age over 18 years who underwent cardiac surgery
Exclusion Criteria
- •data miss greater than 10%
Outcomes
Primary Outcomes
acute kidney injury
Time Frame: 7 days
postoperative AKI was defined according to KDIGO criteria during the first 7 days after operation. Postoperative AKI was defined as either at an increase of at least 50% within 7 days or 0.3 mg/dL elevation within 48 h compared with the reference serum creatinine level.