AutoML Insights into COVID-19 Patient Outcomes in Romania
Author Information
Author(s): Simon Sonja C. S., Bibi Igor, Schaffert Daniel, Benecke Johannes, Martin Niklas, Leipe Jan, Vladescu Cristian, Olsavszky Victor
Primary Institution: University Medical Center and Medical Faculty Mannheim, Heidelberg University
Hypothesis
Can automated machine learning (AutoML) improve the analysis of patient outcomes during the first wave of COVID-19 in Romania?
Conclusion
AutoML provided valuable insights into patient outcomes, highlighting variations in care and the need for targeted health strategies for both COVID-19 and other health challenges.
Supporting Evidence
- Older age, specific hospitals, and oncology wards were less associated with improved recovery rates.
- Mortality was linked to abnormal lab results and cardiovascular/respiratory diseases.
- Patients admitted without referral were more likely to be acute cases.
Takeaway
This study used a computer program to look at a lot of hospital data from COVID-19 patients in Romania to find out how different factors affected their recovery and health.
Methodology
AutoML was used to train and validate models on an ICD-10 dataset covering the first wave of COVID-19 in Romania from January to September 2020.
Potential Biases
The study does not fully account for variations in hospital resource allocation or care protocols, which may have affected the results.
Limitations
The study relies on historical data from a single pandemic wave, which may not reflect outcomes in later waves or different healthcare settings.
Participant Demographics
The dataset included all hospitalized patients in Romania during the specified period, covering various demographics such as age and sex.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website