Table 2 Confusion matrix for predicting presence of PASC among females and males in the test dataset using the stochastic gradient boosting model created from the training dataset

From: Using machine learning involving diagnoses and medications as a risk prediction tool for post-acute sequelae of COVID-19 (PASC) in primary care

	Observed
Predicted	No PASC	PASC	Total
Females (n = 9203)
No PASC	5863	347	6210
PASC	1806	1187	2993
Total	7669	1534	9203
Males (n = 5067)
No PASC	3505	173	3678
PASC	711	678	1398
Total	4216	851	5067

In females, predictions were based on 15,377 trees, with sensitivity 0.774 and specificity 0.765. In males, predictions were based on 11,221 trees, with sensitivity 0.797 and specificity 0.831
PASC, post-acute sequelae of COVID- 19

Back to article page

ISSN: 1741-7015

Contact us

General enquiries: journalsubmissions@springernature.com