Supplemental Digital Content is available in the text.

No abstract

Personal diet management is key to fighting the obesity epidemic. Recent advances in smartphones and wearable sensor technologies have empowered automated food monitoring through food image processing and eating episode detection, with the goal to conquer drawbacks of traditional food journaling that is labour intensive, inaccurate, and low adherent. In this paper, we present a new interactive mobile system that enables automated food recognition and assessment based on user food images and provides dietary intervention while tracking users’ dietary and physical activities. In addition to using techniques in computer vision and machine learning, one unique feature of this system is the realization of real-time energy balance monitoring through metabolic network simulation. As a proof of concept, we have demonstrated the use of this system through an Android application.

Hierarchal clustering of amino acid metabolites may identify a metabolic signature in children with pediatric acute hypoxemic respiratory failure. Seventy-four immunocompetent children, 41 (55.4%) with pediatric acute respiratory distress syndrome (PARDS), who were between 2 days to 18 years of age and within 72 h of intubation for acute hypoxemic respiratory failure, were enrolled. We used hierarchal clustering and partial least squares-discriminant analysis to profile the tracheal aspirate airway fluid using quantitative LC–MS/MS to explore clusters of metabolites that correlated with acute hypoxemia severity and ventilator-free days. Three clusters of children that differed by severity of hypoxemia and ventilator-free days were identified. Quantitative pathway enrichment analysis showed that cysteine and methionine metabolism, selenocompound metabolism, glycine, serine and threonine metabolism, arginine biosynthesis, and valine, leucine, and isoleucine biosynthesis were the top five enriched, impactful pathways. We identified three clusters of amino acid metabolites found in the airway fluid of intubated children important to acute hypoxemia severity that correlated with ventilator-free days < 21 days. Further studies are needed to validate our findings and to test our models.

Background Identifying patterns within ICU medication regimens may help artificial intelligence algorithms to better predict patient outcomes; however, machine learning methods incorporating medications require further development, including standardized terminology. The Common Data Model for Intensive Care Unit (ICU) Medications (CDM-ICURx) may provide important infrastructure to clinicians and researchers to support artificial intelligence analysis of medication-related outcomes and healthcare costs. Using an unsupervised cluster analysis approach in combination with this common data model, the objective of this evaluation was to identify novel patterns of medication clusters (termed ‘pharmacophenotypes’) correlated with ICU adverse events (e.g., fluid overload) and patient-centered outcomes (e.g., mortality). Methods This was a retrospective, observational cohort study of 991 critically ill adults. To identify pharmacophenotypes, unsupervised machine learning analysis with automated feature learning using restricted Boltzmann machine and hierarchical clustering was performed on the medication administration records of each patient during the first 24 h of their ICU stay. Hierarchical agglomerative clustering was applied to identify unique patient clusters. Distributions of medications across pharmacophenotypes were described, and differences among patient clusters were compared using signed rank tests and Fisher's exact tests, as appropriate. Results A total of 30,550 medication orders for the 991 patients were analyzed; five unique patient clusters and six unique pharmacophenotypes were identified. For patient outcomes, compared to patients in Clusters 1 and 3, patients in Cluster 5 had a significantly shorter duration of mechanical ventilation and ICU length of stay (p < 0.05); for medications, Cluster 5 had a higher distribution of Pharmacophenotype 1 and a smaller distribution of Pharmacophenotype 2, compared to Clusters 1 and 3. For outcomes, patients in Cluster 2, despite having the highest severity of illness and greatest medication regimen complexity, had the lowest overall mortality; for medications, Cluster 2 also had a comparably higher distribution of Pharmacophenotype 6. Conclusion The results of this evaluation suggest that patterns among patient clusters and medication regimens may be observed using empiric methods of unsupervised machine learning in combination with a common data model. These results have potential because while phenotyping approaches have been used to classify heterogenous syndromes in critical illness to better define treatment response, the entire medication administration record has not been incorporated in those analyses. Applying knowledge of these patterns at the bedside requires further algorithm development and clinical application but may have the future potential to be leveraged in guiding medication-related decision making to improve treatment outcomes.

Children with life-threatening asthma exacerbations who are admitted to a pediatric intensive care unit (PICU) are a heterogeneous group with poorly studied inflammatory features. We hypothesized that distinct clusters of children with asthma in a PICU would be identified based on differences in plasma cytokine levels and that these clusters would have differing underlying inflammation and asthma outcomes within 1 year. Plasma cytokines and differential gene expression were measured in neutrophils isolated from children admitted to a PICU for asthma. Participants were clustered by differential plasma cytokine abundance. Gene expression differences were compared by cluster and pathway over-representation analysis was performed. We identified two clusters in 69 children with no clinical differences. Cluster 1 (n = 41) had higher cytokines compared to Cluster 2 (n = 28). Cluster 2 had a hazard ratio of 2.71 (95% CI 1.11–6.64) compared to Cluster 1 for time to subsequent exacerbation. Gene expression pathways that differed by cluster included interleukin-10 signaling; nucleotide-binding domain, leucine rich repeat containing receptor (NLR signaling); and toll-like receptor (TLR) signaling. These observations suggest that a subset of children may have a unique pattern of inflammation during PICU hospitalization that might require alternative treatment approaches.

Background/Objectives: Eosinophilic esophagitis (EoE) is an inflammatory disease of unclear etiology. The aim of this study was to use untargeted plasma metabolomics to identify metabolic pathway alterations associated with EoE to better understand the pathophysiology. Methods: This prospective, case-control study included 72 children, aged 1–17 years, undergoing clinically indicated upper endoscopy (14 diagnosed with EoE and 58 controls). Fasting plasma samples were analyzed for metabolomics by high-resolution dual-chromatography mass spectrometry. Analysis was performed on sex-matched groups at a 2:1 ratio. Significant differences among the plasma metabolite features between children with and without EoE were determined using multivariate regression analysis and were annotated with a network-based algorithm. Subsequent pathway enrichment analysis was performed. Results: Patients with EoE had a higher proportion of atopic disease (85.7% vs 50%, P = 0.019) and any allergies (100% vs 57.1%, P = 0.0005). Analysis of the dual chromatography features resulted in a total of 918 metabolites that differentiated EoE and controls. Glycerophospholipid metabolism was significantly enriched with the greatest number of differentiating metabolites and overall pathway enrichment (P < 0.01). Multiple amino and fatty acid pathways including linoleic acid were also enriched, as well as pyridoxine metabolism (P < 0.01). Conclusions: In this pilot study, we found differences in metabolites involved in glycerophospholipid and inflammation pathways in pediatric patients with EoE using untargeted metabolomics, as well as overlap with amino acid metabolome alterations found in atopic disease.

The host immune response to a viral immune stimulus has not been examined in children during a life-threatening asthma attack. We determined whether we could identify clusters of children with critical asthma by functional immunophenotyping using an intracellular viral analog stimulus. We performed a single-center, prospective, observational cohort study of 43 children ages 6–17 years admitted to a pediatric intensive care unit for an asthma attack between July 2019 to February 2021. Neutrophils were isolated from children, stimulated overnight with LyoVec poly(I:C), and mRNA was analyzed using a targeted Nanostring immunology array. Network analysis of the differentially expressed transcripts for the paired LyoVec poly(I:C) samples was performed. We identified two clusters by functional immunophenotyping that differed by the Asthma Control Test score. Cluster 1 (n = 23) had a higher proportion of children with uncontrolled asthma in the four weeks prior to PICU admission compared with cluster 2 (n = 20). Pathways up-regulated in cluster 1 versus cluster 2 included chemokine receptor/chemokines, interleukin-10 (IL-10), IL-4, and IL-13 signaling. Larger validation studies and clinical phenotyping of children with critical asthma are needed to determine the predictive utility of these clusters in a larger clinical setting.