Search

Preterm infants are prone to NeuroDevelopmental Impairment (NDI). Some previous works have identified clinical variables that can be potential predictors of NDI. However, machine learning (ML)-based models still present low predictive capabilities when addressing this problem. This work attempts to evaluate the application of ML techniques to predict NDI using clinical data from a cohort of very preterm infants recruited at birth and assessed at 2 years of age. Six different classification models were assessed, using all features, clinician-selected features, and mutual information feature selection. The best results were obtained by ML models trained using mutual information-selected features and employing oversampling, for cognitive and motor impairment prediction, while for language impairment prediction the best setting was clinician-selected features. Although the performance indicators in this local cohort are consistent with similar previous works and still rather poor. This is a clear indication that, in order to obtain better performance rates, further analysis and methods should be considered, and other types of data should be taken into account together with the clinical variables.

The role of rhinoviruses (RVs) in children with clinical syndromes not classically associated with RV infections is not well understood. We analyzed a cohort of children ≤21 years old who were PCR+ for RV at a large Pediatric Hospital from 2011 to 2013. Using univariate and multivariable logistic regression, we analyzed the associations between demographic, clinical characteristics, microbiology data, and clinical outcomes in children with compatible symptoms and incidental RV detection. Of the 2473 children (inpatients and outpatients) with an RV+ PCR, 2382 (96%) had compatible symptoms, and 91 (4%) did not. The overall median age was 14 months and 78% had underlying comorbidities. No differences in RV viral loads were found according to the presence of compatible symptoms, while in children with classic RV symptoms, RV viral loads were higher in single RV infections versus RV viral co-infections. Bacterial co-infections were more common in RV incidental detection (7.6%) than in children with compatible symptoms (1.9%, p < 0.001). The presence of compatible symptoms independently increased the odds ratio (OR, 95% CI) of hospitalization 4.8 (3.1–7.4), prolonged hospital stays 1.9 (1.1–3.1), need for oxygen 12 (5.8–25.0) and pediatric intensive care unit (PICU) admission 4.13 (2.0–8.2). Thus, despite comparable RV loads, disease severity was significantly worse in children with compatible symptoms.