Back

Switching from algorithm-based to universal admission screening for COVID-19 in hospital settings

Donker, T.; Grundmann, H.; Bürkin, F.; Hengel, H.; Bürkle, H.; Hammer, T.; Wenz, F.; Kern, W.

2020-08-11 infectious diseases
10.1101/2020.08.07.20170001 medRxiv
Show abstract

Detection of COVID-19 positive cases on admission to hospitals is crucial to protect patients and staff at the same time. While universal admission screening can prevent more undetected introductions than the algorithm-based screening, which preselects patient based on their symptoms and exposure, it is a more costly strategy as it involves testing a large number of patients. We construct a simple tool to help determine when the benefit of additionally found cases outweighs the cost of the additionally tested patients, based on the numbers of patients to be screened in an acceptable time span to find an additional case when screening all admitted patients.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.