Abstract
Differential diagnosis between bacterial and viral meningitis is crucial. In our study, to differentiate bacterial vs. viral meningitis, three machine learning (ML) algorithms (multiple logistic regression (MLR), random forest (RF), and naïve-Bayes (NB)) were applied for the two age groups (0–14 and >14 years) of patients with meningitis by both conventional (culture) and molecular (PCR) methods. Cerebrospinal fluid (CSF) neutrophils, CSF lymphocytes, neutrophil-to-lymphocyte ratio (NLR), blood albumin, blood C-reactive protein (CRP), glucose, blood soluble urokinase-type plasminogen activator receptor (suPAR), and CSF lymphocytes-to-blood CRP ratio (LCR) were used as predictors for the ML algorithms. The performance of the ML algorithms was evaluated through a cross-validation procedure, and optimal predictions of the type of meningitis were above 95% for viral and 78% for bacterial meningitis. Overall, MLR and RF yielded the best performance when using CSF neutrophils, CSF lymphocytes, NLR, albumin, glucose, gender, and CRP. Also, our results reconfirm the high diagnostic accuracy of NLR in the differential diagnosis between bacterial and viral meningitis.
Highlights
Mitigating meningitis remains both a global health challenge and a clinical emergency issue, the latter even in resource-rich settings [1,2]
In doing so, using cerebrospinal fluid (CSF) neutrophils, CSF lymphocytes, neutrophil-to-lymphocyte ratio (NLR), blood albumin, blood glucose, gender, blood C-reactive protein (CRP), and blood soluble urokinase-type plasminogen activator receptor (suPAR) as predictors, we identified 99 available cases distributed among 59 (60%) viral and 40 (40%) bacterial cases with a training set of 79 cases, and testing set of 20 cases (Table 3f)
Using CSF neutrophils, CSF lymphocytes, NLR, blood albumin, blood glucose, gender, blood CRP, blood suPAR, and lymphocytes-to-blood CRP ratio (LCR) as predictors, we identified 99 available cases distributed among 59 (60%) viral and 40 (40%) bacterial cases
Summary
Mitigating meningitis remains both a global health challenge and a clinical emergency issue, the latter even in resource-rich settings [1,2]. Of paramount importance is its prompt diagnosis and, in particular, the differential diagnosis between the two main categories, bacterial and viral meningitis [3]. The latter is crucial for two main reasons: (a) failure to deliver proper antibiotic therapy in bacterial meningitis can lead to severe, permanent sequelae and invasive disease (especially due to Neisseria meningitidis) [4], and even death, and (b) unnecessary antibiotic or overtreatment of viral meningitis cases can lead to antimicrobial resistance, increased health care services cost, changes in human microbiome, and high levels of stress to the suffering patients [5]. To our best knowledge, the combined effects with other CSF and blood parameters, beyond those referring to whole white cell counts, lymphocytes, and neutrophils, have not been explored in the differential diagnosis between bacterial and viral meningitis. We employed three different ML algorithms, and we found that the accuracy in the differential diagnosis of meningitis might be increased when these algorithms are used in a multivariate approach, instead of a ROC curve univariate treatment of the problem
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
