A deep learning model for predicting multidrug-resistant organism infection in critically ill patients
Authors' conclusion
This study confirmed nine independent risk factors for MDRO infection. The BPNN model performed well and was potentially used to predict MDRO infection in ICU patients.
Study context
- Population
- Critically ill patients admitted to the Intensive Care Unit (ICU) of the Affiliated Hospital of Qingdao University, August 2021 to January 2022primary cohort
- Participants
- 688 patients
- Comparison
- Training set vs Test set vs Validation set
- Validation setn = 238
- Outcome
- Multidrug-resistant organism (MDRO) infection prediction (AUC)
Methodological detailssingle-centre, random training/test split (80%/20%), external (temporal) validation, backpropagation neural network
- Single-centre
- Random training/test split (80%/20%)
- External (temporal) validation
- Backpropagation neural network
Performance
As reported in the abstract
AUC of backpropagation neural network (BPNN) model for MDRO infection
| Test or model | AUC |
|---|---|
| Training set | 0.889 (95% CI 0.852-0.925) |
| Test set | 0.919 (95% CI 0.856-0.983) |
| Validation set | 0.811 (95% CI 0.731-0.891) |
AUC
Training set0.889 (95% CI 0.852-0.925)
Test set0.919 (95% CI 0.856-0.983)
Validation set0.811 (95% CI 0.731-0.891)
Further findings
| Outcome | Validation set |
|---|---|
| MDRO infection incidence | 13.03% |
- Multidrug-resistant organism (MDRO) infection incidence: Primary cohort 15.84%
Independent risk factors for MDRO infection
- Length of hospitalization, ICU LOS, long-term bed rest, antibiotics use before ICU, APACHE II, invasive operation before ICU, quantity of antibiotics, chronic lung disease, hypoproteinemia
- 9
Original publication
A deep learning model for predicting multidrug-resistant organism infection in critically ill patients. J Intensive Care. Published 9 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →