Cardiothoracic & vascular anesthesia
Predicting In-Hospital Acute Kidney Injury after Cardiac Surgery Using Machine Learning
Authors' conclusion
We developed two machine learning models that incorporated perioperative biomarkers and hemodynamics as input data to predict CSA-AKI. The models exhibited competitive performance for prediction of the outcome
Study context
- Population
- Adults undergoing elective and urgent cardiac surgery, single-centre university hospital, May 2015–March 2023602 included in models
- Participants
- Six hundred sixty-seven adults
- Exposure
- Random forest (RF) vs long short-term memory (LSTM) machine learning models
- Outcome
- Cardiac surgery-associated acute kidney injury (CSA-AKI) prediction, test set (n=110)
Methodological detailsmachine learning prediction model development, single-centre
- Machine learning prediction model development
- Single-centre
Main finding
As reported in the abstract
| Group | AUC | Sensitivity | Specificity | Accuracy |
|---|---|---|---|---|
| RF | 0.73 (0.72-0.74) | 0.59 (0.57-0.62) | 0.81 (0.79-0.84) | 0.77 (0.76-0.79) |
| LSTM | 0.68 (0.67-0.70) | 0.60 (0.57-0.64) | 0.74 (0.71-0.78) | 0.72 (0.69-0.74) |
AUC
RF0.73 (0.72-0.74)
LSTM0.68 (0.67-0.70)
Sensitivity
RF0.59 (0.57-0.62)
LSTM0.60 (0.57-0.64)
Specificity
RF0.81 (0.79-0.84)
LSTM0.74 (0.71-0.78)
Accuracy
RF0.77 (0.76-0.79)
LSTM0.72 (0.69-0.74)
Observational data: associations, not effects.
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
Predicting In-Hospital Acute Kidney Injury after Cardiac Surgery Using Machine Learning. J Cardiothorac Vasc Anesth. Published 23 September 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →