Background The rapid development in big data analytics and the data-rich environment of intensive care units together provide unprecedented opportunities for medical breakthroughs in the field of critical care. We developed and validated a machine learning-based model, the Pediatric Risk of Mortality Prediction Tool (PROMPT), for real-time prediction of all-cause mortality in pediatric intensive care units. Methods Utilizing two separate retrospective observational cohorts, we conducted model development and validation using a machine learning algorithm with a convolutional neural network. The development cohort comprised 1445 pediatric patients with 1977 medical encounters admitted to intensive care units from January 2011 to December 2017 at Severance Hospital (Seoul, Korea). The validation cohort included 278 patients with 364 medical encounters admitted to the pediatric intensive care unit from January 2016 to November 2017 at Samsung Medical Center. Results Using seven vital signs, along with patient age and body weight on intensive care unit admission, PROMPT achieved an area under the receiver operating characteristic curve in the range of 0.89–0.97 for mortality prediction 6 to 60 h prior to death. Our results demonstrated that PROMPT provided high sensitivity with specificity and outperformed the conventional severity scoring system, the Pediatric Index of Mortality, in predictive ability. Model performance was indistinguishable between the development and validation cohorts. Conclusions PROMPT is a deep model-based, data-driven early warning score tool that can predict mortality in critically ill children and may be useful for the timely identification of deteriorating patients. Electronic supplementary material The online version of this article (10.1186/s13054-019-2561-z) contains supplementary material, which is available to authorized users.

BackgroundThe number of children using home mechanical ventilation (HMV) has increased markedly in Europe and North America, but little is known about the situation in Korea. We described the clinical characteristics of children using HMV and investigated the current situation of HMV utilization in children.MethodsData on HMV prescriptions in year 2016 for children under the age of 19 was retrieved from the National Health Insurance Service for nationwide information. For more detailed information, data from year 2016 to 2018 was also retrieved from a tertiary center, Severance Children's Hospital.ResultsNationwide, 416 children were prescribed with HMV in 2016, with an estimated prevalence of 4.4 per 100,000 children, of which 64.2% were male and mean age was 6-year-old. The estimated number of patients using invasive ventilators via tracheostomy was 202 (49%). Neuromuscular diseases were the most frequent cause (217; 52%), followed by central nervous system diseases (142; 34%), and cardiopulmonary diseases (57; 14%). In the tertiary center, a total of 62 children were prescribed with HMV (19 [31%] with non-invasive ventilation; 43 [69%] with invasive ventilation]. The number of children with HMV increased from 11 in 2016 to 29 in 2018. The mean age for initiation of HMV was 3.1 years and male patients comprised 65%. The most frequent diagnostic reason for HMV was central nervous system diseases (68%), followed by cardiopulmonary diseases (19%) and neuromuscular diseases (13%). Five patients died during the study period and five patients weaned from HMV.ConclusionThis study provides insights on the present situation of HMV utilization in Korean children.

No abstract

The mitochondrial genome encodes core catalytic peptides that affect major metabolic processes within a cell. Here, we investigated the association between mitochondrial DNA (mtDNA) variants and allergic diseases, including atopic dermatitis (AD) and asthma, alongside heteroplasmy within the mtDNA in subjects with allergic sensitization. We collected genotype data from 973 subjects with allergic sensitization, consisting of 632 children with AD, 498 children with asthma, and 481 healthy controls by extracting DNA from their blood samples. Fisher's exact test was used to investigate mtDNA and nuclear DNA variants related to mitochondrial function (MT-nDNA) to identify their association with allergic diseases. Among the 69 mtDNA variants, rs28357671 located on the MT-ND6 gene displayed statistically significant associations with allergic diseases (Bonferroni-corrected P < 7.25E-4), while 6, 4, and 2 genes were associated with allergic sensitization, AD, and asthma, respectively ( P < 0.0002), including NLRX1 , OCA2 , and CHCHD3 among the MT-nDNA genes. Heteroplasmy of mitochondrial DNA associated with allergic sensitization was evaluated in a separate cohort of patients consisting of 59 subjects with allergic sensitization and 52 controls. Heteroplasmy was verified when a patient carried both alleles of a mitochondrial single-nucleotide polymorphism (SNP) when clustered. One of the 134 mitochondrial SNPs showed heteroplasmy at a level of 0.4313 when clustering was applied. The probe sequence located at mitochondrial position 16217 and within the D-loop, which acts as a major control site for mtDNA expression. This is the first study to evaluate the association between mitochondrial DNA variants and allergic diseases. A harmonized effect of genes related to mitochondrial function may contribute to the risk of allergic diseases.