Highly comparative time series analysis of oxygen saturation and heart rate to predict respiratory outcomes in extremely preterm infants.

OBJECTIVE: Highly comparative time series analysis (HCTSA) is a novel approach involving massive feature extraction using publicly available code from many disciplines. The Prematurity-Related Ventilatory Control (Pre-Vent) observational multicenter prospective study collected bedside monitor data from >700 extremely preterm infants to identify physiologic features that predict respiratory outcomes. We calculated a subset of 33 HCTSA features on >7M 10-minute windows of oxygen saturation (SPO2) and heart rate (HR) from the PreVent cohort to quantify predictive performance. This subset included representatives previously identified using unsupervised clustering on >3500 HCTSA algorithms. Performance of each feature was measured by individual area under the receiver operating curve (AUC) at various days of life and binary respiratory outcomes. We hypothesized that the best HCTSA algorithms would compare favorably to optimal PreVent physiologic predictor IH90_DPE (duration per event of intermittent hypoxemia events below 90%). Main Results: The top HCTSA features were from a cluster of algorithms associated with the autocorrelation of SPO2 time series and identified low frequency patterns of desaturation as high risk. These features had comparable performance to and were highly correlated with IH90_DPE but perhaps measure the physiologic status of an infant in a more robust way that warrants further investigation. The top HR HCTSA features were symbolic transformation measures that had previously been identified as strong predictors of neonatal mortality. HR metrics were only important predictors at early days of life which was likely due to the larger proportion of infants whose outcome was death by any cause. A simple HCTSA model using 3 top features outperformed IH90_DPE at day of life 7 (.778 versus .729) but was essentially equivalent at day of life 28 (.849 versus .850). These results validated the utility of a representative HCTSA approach but also provides additional evidence supporting IH90\_DPE as an optimal predictor of respiratory outcomes.

Apnea, Intermittent Hypoxemia, and Bradycardia Events Predict Late-Onset Sepsis in Infants Born Extremely Preterm.

OBJECTIVE: The objective of this study was to examine the association of cardiorespiratory events, including apnea, periodic breathing, intermittent hypoxemia (IH), and bradycardia, with late-onset sepsis for extremely preterm infants (<29 weeks of gestational age) on vs off invasive mechanical ventilation. STUDY DESIGN: This is a retrospective analysis of data from infants enrolled in Pre-Vent (ClinicalTrials.gov identifier NCT03174301), an observational study in 5 level IV neonatal intensive care units. Clinical data were analyzed for 737 infants (mean gestational age: 26.4 weeks, SD 1.71). Monitoring data were available and analyzed for 719 infants (47 512 patient-days); of whom, 109 had 123 sepsis events. Using continuous monitoring data, we quantified apnea, periodic breathing, bradycardia, and IH. We analyzed the relationships between these daily measures and late-onset sepsis (positive blood culture >72 hours after birth and ≥5-day antibiotics). RESULTS: For infants not on a ventilator, apnea, periodic breathing, and bradycardia increased before sepsis diagnosis. During times on a ventilator, increased sepsis risk was associated with longer events with oxygen saturation <80% (IH80) and more bradycardia events before sepsis. IH events were associated with higher sepsis risk but did not dynamically increase before sepsis, regardless of ventilator status. A multivariable model including postmenstrual age, cardiorespiratory variables (apnea, periodic breathing, IH80, and bradycardia), and ventilator status predicted sepsis with an area under the receiver operator characteristic curve of 0.783. CONCLUSION: We identified cardiorespiratory signatures of late-onset sepsis. Longer IH events were associated with increased sepsis risk but did not change temporally near diagnosis. Increases in bradycardia, apnea, and periodic breathing preceded the clinical diagnosis of sepsis.

Apnea, Intermittent Hypoxemia, and Bradycardia Events Predict Late-Onset Sepsis in Extremely Preterm Infants.

Objectives: Detection of changes in cardiorespiratory events, including apnea, periodic breathing, intermittent hypoxemia (IH), and bradycardia, may facilitate earlier detection of sepsis. Our objective was to examine the association of cardiorespiratory events with late-onset sepsis for extremely preterm infants (<29 weeks' gestational age (GA)) on versus off invasive mechanical ventilation. Study Design: Retrospective analysis of data from infants enrolled in Pre-Vent (ClinicalTrials.gov identifier NCT03174301), an observational study in five level IV neonatal intensive care units. Clinical data were analyzed for 737 infants (mean GA 26.4w, SD 1.71). Monitoring data were available and analyzed for 719 infants (47,512 patient-days), of whom 109 had 123 sepsis events. Using continuous monitoring data, we quantified apnea, periodic breathing, bradycardia, and IH. We analyzed the relationships between these daily measures and late-onset sepsis (positive blood culture >72h after birth and ≥5d antibiotics). Results: For infants not on a ventilator, apnea, periodic breathing, and bradycardia increased before sepsis diagnosis. During times on a ventilator, increased sepsis risk was associated with longer IH80 events and more bradycardia events before sepsis. IH events were associated with higher sepsis risk, but did not dynamically increase before sepsis, regardless of ventilator status. A multivariable model predicted sepsis with an AUC of 0.783. Conclusion: We identified cardiorespiratory signatures of late-onset sepsis. Longer IH events were associated with increased sepsis risk but did not change temporally near diagnosis. Increases in bradycardia, apnea, and periodic breathing preceded the clinical diagnosis of sepsis.

Highly comparative time series analysis of oxygen saturation and heart rate to predict respiratory outcomes in extremely preterm infants.

Objective: Highly comparative time series analysis (HCTSA) is a novel approach involving massive feature extraction using publicly available code from many disciplines. The Prematurity-Related Ventilatory Control (Pre-Vent) observational multicenter prospective study collected bedside monitor data from > 700 extremely preterm infants to identify physiologic features that predict respiratory outcomes. We calculated a subset of 33 HCTSA features on > 7M 10-minute windows of oxygen saturation (SPO2) and heart rate (HR) from the Pre-Vent cohort to quantify predictive performance. This subset included representatives previously identified using unsupervised clustering on > 3500 HCTSA algorithms. Performance of each feature was measured by individual area under the receiver operating curve (AUC) at various days of life and binary respiratory outcomes. These were compared to optimal PreVent physiologic predictor IH90 DPE, the duration per event of intermittent hypoxemia events with threshold of 90%. Main Results: The top HCTSA features were from a cluster of algorithms associated with the autocorrelation of SPO2 time series and identified low frequency patterns of desaturation as high risk. These features had comparable performance to and were highly correlated with IH90_DPE but perhaps measure the physiologic status of an infant in a more robust way that warrants further investigation. The top HR HCTSA features were symbolic transformation measures that had previously been identified as strong predictors of neonatal mortality. HR metrics were only important predictors at early days of life which was likely due to the larger proportion of infants whose outcome was death by any cause. A simple HCTSA model using 3 top features outperformed IH90_DPE at day of life 7 (.778 versus .729) but was essentially equivalent at day of life 28 (.849 versus .850). These results validated the utility of a representative HCTSA approach but also provides additional evidence supporting IH90_DPE as an optimal predictor of respiratory outcomes.

Access to healthcare and depression severity in vulnerable groups the US: NHANES 2013-2018.

BACKGROUND: Access to healthcare is essential for managing chronic diseases, yet it often poses a barrier, contributing to a significant burden of conditions like depression. This study aimed to investigate the association between healthcare access and depression severity in contemporary free-living adults in the US, with a focus on identifying vulnerable populations. METHOD: Data from the National Health and Nutrition Examination Survey cycles 2013-2018 were utilized, involving 13,689 participants aged 20 years or older. Multivariable multinomial logistic regression models were conducted, adjusting for various confounding variables. RESULTS: Approximately 17 % of US adults lacked access to healthcare, while 24 % experienced varying levels of depression severity, with 8 % having moderate-to-severe depression. More males faced challenges accessing healthcare, while more females reported diverse levels of depression. Both healthcare access and depression severity were associated with low educational attainment, low familial income, lacking spousal support, lacking health insurance coverage, and worse self-reported overall health. We found a higher vulnerability to moderate-to-severe depression among females (OR (95 % CI): 1.20 (0.91, 1.59)), individuals identifying as the Other ethnic group (1.69 (1.02, 2.79)), and those living without a spouse (1.57 (1.10, 2.26)). LIMITATIONS: Our cross-sectional study cannot establish causality, and potential biases related to self-reported data exist. CONCLUSIONS: Access to healthcare emerged as a crucial predictor of moderate-to-severe depression among females, individuals of the Other ethnic group, and those without a spouse. Longitudinal research is needed to confirm and enhance our understanding of factors that shape the relationship between healthcare access and depression in free-living US adults.

Graphical Association Analysis for Identifying Variation in Provider Claims for Joint Replacement Surgery.

Identifying potentially fraudulent or wasteful medical insurance claims can be difficult due to the large amounts of data and human effort involved. We applied unsupervised machine learning to construct interpretable models which rank variations in medical provider claiming behaviour in the domain of unilateral joint replacement surgery, using data from the Australian Medicare Benefits Schedule. For each of three surgical procedures reference models of claims for each procedure were constructed and compared analytically to models of individual provider claims. Providers were ranked using a score based on fees for typical claims made in addition to those in the reference model. Evaluation of the results indicated that the top-ranked providers were likely to be unusual in their claiming patterns, with typical claims from outlying providers adding up to 192% to the cost of a procedure. The method is efficient, generalizable to other procedures and, being interpretable, integrates well into existing workflows.

Maturation of cardioventilatory physiological trajectories in extremely preterm infants.

BACKGROUND: In extremely preterm infants, persistence of cardioventilatory events is associated with long-term morbidity. Therefore, the objective was to characterize physiologic growth curves of apnea, periodic breathing, intermittent hypoxemia, and bradycardia in extremely preterm infants during the first few months of life. METHODS: The Prematurity-Related Ventilatory Control study included 717 preterm infants <29 weeks gestation. Waveforms were downloaded from bedside monitors with a novel sharing analytics strategy utilized to run software locally, with summary data sent to the Data Coordinating Center for compilation. RESULTS: Apnea, periodic breathing, and intermittent hypoxemia events rose from day 3 of life then fell to near-resolution by 8-12 weeks of age. Apnea/intermittent hypoxemia were inversely correlated with gestational age, peaking at 3-4 weeks of age. Periodic breathing was positively correlated with gestational age peaking at 31-33 weeks postmenstrual age. Females had more periodic breathing but less intermittent hypoxemia/bradycardia. White infants had more apnea/periodic breathing/intermittent hypoxemia. Infants never receiving mechanical ventilation followed similar postnatal trajectories but with less apnea and intermittent hypoxemia, and more periodic breathing. CONCLUSIONS: Cardioventilatory events peak during the first month of life but the actual postnatal trajectory is dependent on the type of event, race, sex and use of mechanical ventilation. IMPACT: Physiologic curves of cardiorespiratory events in extremely preterm-born infants offer (1) objective measures to assess individual patient courses and (2) guides for research into control of ventilation, biomarkers and outcomes. Presented are updated maturational trajectories of apnea, periodic breathing, intermittent hypoxemia, and bradycardia in 717 infants born <29 weeks gestation from the multi-site NHLBI-funded Pre-Vent study. Cardioventilatory events peak during the first month of life but the actual postnatal trajectory is dependent on the type of event, race, sex and use of mechanical ventilation. Different time courses for apnea and periodic breathing suggest different maturational mechanisms.

Histone locus bodies: a paradigm for how nuclear biomolecular condensates control cell cycle regulated gene expression.

Histone locus bodies (HLBs) are biomolecular condensates that assemble at replication-dependent (RD) histone genes in animal cells. These genes produce unique mRNAs that are not polyadenylated and instead end in a conserved 3' stem loop critical for coordinated production of histone proteins during S phase of the cell cycle. Several evolutionarily conserved factors necessary for synthesis of RD histone mRNAs concentrate only in the HLB. Moreover, because HLBs are present throughout the cell cycle even though RD histone genes are only expressed during S phase, changes in HLB composition during cell cycle progression drive much of the cell cycle regulation of RD histone gene expression. Thus, HLBs provide a powerful opportunity to determine the cause-and-effect relationships between nuclear body formation and cell cycle regulated gene expression. In this review, we focus on progress during the last five years that has advanced our understanding of HLB biology.

A multi-taxon analysis of European Red Lists reveals major threats to biodiversity.

Biodiversity loss is a major global challenge and minimizing extinction rates is the goal of several multilateral environmental agreements. Policy decisions require comprehensive, spatially explicit information on species' distributions and threats. We present an analysis of the conservation status of 14,669 European terrestrial, freshwater and marine species (ca. 10% of the continental fauna and flora), including all vertebrates and selected groups of invertebrates and plants. Our results reveal that 19% of European species are threatened with extinction, with higher extinction risks for plants (27%) and invertebrates (24%) compared to vertebrates (18%). These numbers exceed recent IPBES (Intergovernmental Platform on Biodiversity and Ecosystem Services) assumptions of extinction risk. Changes in agricultural practices and associated habitat loss, overharvesting, pollution and development are major threats to biodiversity. Maintaining and restoring sustainable land and water use practices is crucial to minimize future biodiversity declines.

Cardiorespiratory Monitoring Data to Predict Respiratory Outcomes in Extremely Preterm Infants.

Rationale: Immature control of breathing is associated with apnea, periodic breathing, intermittent hypoxemia, and bradycardia in extremely preterm infants. However, it is not clear if such events independently predict worse respiratory outcome. Objectives: To determine if analysis of cardiorespiratory monitoring data can predict unfavorable respiratory outcomes at 40 weeks postmenstrual age (PMA) and other outcomes, such as bronchopulmonary dysplasia at 36 weeks PMA. Methods: The Prematurity-related Ventilatory Control (Pre-Vent) study was an observational multicenter prospective cohort study including infants born at <29 weeks of gestation with continuous cardiorespiratory monitoring. The primary outcome was either "favorable" (alive and previously discharged or inpatient and off respiratory medications/O2/support at 40 wk PMA) or "unfavorable" (either deceased or inpatient/previously discharged on respiratory medications/O2/support at 40 wk PMA). Measurements and Main Results: A total of 717 infants were evaluated (median birth weight, 850 g; gestation, 26.4 wk), 53.7% of whom had a favorable outcome and 46.3% of whom had an unfavorable outcome. Physiologic data predicted unfavorable outcome, with accuracy improving with advancing age (area under the curve, 0.79 at Day 7, 0.85 at Day 28 and 32 wk PMA). The physiologic variable that contributed most to prediction was intermittent hypoxemia with oxygen saturation as measured by pulse oximetry <90%. Models with clinical data alone or combining physiologic and clinical data also had good accuracy, with areas under the curve of 0.84-0.85 at Days 7 and 14 and 0.86-0.88 at Day 28 and 32 weeks PMA. Intermittent hypoxemia with oxygen saturation as measured by pulse oximetry <80% was the major physiologic predictor of severe bronchopulmonary dysplasia and death or mechanical ventilation at 40 weeks PMA. Conclusions: Physiologic data are independently associated with unfavorable respiratory outcome in extremely preterm infants.

Quantitative and Qualitative Changes in Peripheral Chemoreceptor Activity in Preterm Infants.

Rationale: Preterm infants are at risk for ventilatory control instability that may be due to aberrant peripheral chemoreceptor activity. Although term infants have increasing peripheral chemoreceptor contribution to overall ventilatory drive with increasing postnatal age, how peripheral chemoreceptor contribution changes in preterm infants with increasing postmenstrual age is not known. Objectives: To evaluate peripheral chemoreceptor activity between 32 and 52 weeks postmenstrual age in preterm infants, using both quantitative and qualitative measures. Methods: Fifty-five infants born between 24 weeks, 0 days gestation and 28 weeks, 6 days gestation underwent hyperoxic testing at one to four time points between 32 and 52 weeks postmenstrual age. Quantitative [Formula: see text] decreases were calculated, and qualitative responses were categorized as apnea, continued breathing with a clear reduction in [Formula: see text], sigh breaths, and no response. Measurements and Main Results: A total of 280 hyperoxic tests were analyzed (2.2 ± 0.3 tests per infant at each time point). Mean peripheral chemoreceptor contribution to ventilatory drive was 85.2 ± 20.0% at 32 weeks and 64.1 ± 22.0% at 52 weeks. Apneic responses were more frequent at earlier postmenstrual ages. Conclusions: Among preterm infants, the peripheral chemoreceptor contribution to ventilatory drive was greater at earlier postmenstrual ages. Apnea was a frequent response to hyperoxic testing at earlier postmenstrual ages, suggesting high peripheral chemoreceptor activity. A clearer description of how peripheral chemoreceptor activity changes over time in preterm infants may help explain how ventilatory control instability contributes to apnea and sleep-disordered breathing later in childhood. Clinical trial registered with www.clinicaltrials.gov (NCT03464396).

Coordinating transcription and replication to mitigate their conflicts in early Drosophila embryos.

Collisions between transcribing RNA polymerases and DNA replication forks are disruptive. The threat of collisions is particularly acute during the rapid early embryonic cell cycles of Drosophila when S phase occupies the entirety of interphase. We hypothesize that collision-avoidance mechanisms safeguard this early transcription. Real-time imaging of endogenously tagged RNA polymerase II (RNAPII) and a reporter for nascent transcripts in unperturbed embryos shows clustering of RNAPII at around 2 min after mitotic exit, followed by progressive dispersal as associated nascent transcripts accumulate later in interphase. Abrupt inhibition of various steps in DNA replication, including origin licensing, origin firing, and polymerization, suppresses post-mitotic RNAPII clustering and transcription in nuclear cycles. We propose that replication dependency defers the onset of transcription so that RNAPII transcribes behind advancing replication forks. The resulting orderly progression can explain how early embryos circumvent transcription-replication conflicts to express essential developmental genes.

Pulmonary function tests in extremely low gestational age infants at one year of age.

RATIONALE: Identifying neonatal and post-discharge exposures among extremely low gestational age newborns (ELGANs) that drive increased pulmonary morbidity and abnormal lung function at 1 year of age proves challenging. OBJECTIVE: The NIH-sponsored Prematurity and Respiratory Outcomes Program (PROP), evaluated infant pulmonary function tests (iPFTs) at 1 year corrected age to determine which demographic and clinical factors are associated with abnormal lung function. METHODS: iPFTs were performed on a PROP subcohort of 135 participants following Institutional Review Board (IRB)-approved written consent. Demographic data, Neonatal Intensive Care Unit (NICU) clinical care, and post-NICU exposures were analyzed for association with iPFTs. MAIN RESULTS: A significant decrease in forced expiratory volume at 0.5 s (FEV0.5 ) and/or forced expiratory flows at 75% of forced vital capacity (FEF75 ), were associated with male sex and African American race. Clinical factors including longer duration of ventilatory support, exposure to systemic steroids, and weight less than the 10th percentile at 36 weeks postmenstrual age were also associated with airflow obstruction, whereas supplemental oxygen requirement and bronchopulmonary dysplasia were not. Additionally, the need for respiratory medications, technology, or hospitalizations during the first year, ascertained by a quarterly survey, were the only post-NICU factors associated with decreased FEV0.5 and FEF75 . Only 7% of infants had reversible airflow obstruction. CONCLUSIONS: Neonatal demographic factors, respiratory support in the NICU, and a history of greater post-NICU medical utilization for respiratory disease had the strongest association with lower lung function at 1 year in ELGANs.

Persistence of Multiple Paramyxoviruses in a Closed Captive Colony of Fruit Bats (Eidolon helvum).

Bats have been identified as the natural hosts of several emerging zoonotic viruses, including paramyxoviruses, such as Hendra and Nipah viruses, that can cause fatal disease in humans. Recently, African fruit bats with populations that roost in or near urban areas have been shown to harbour a great diversity of paramyxoviruses, posing potential spillover risks to public health. Understanding the circulation of these viruses in their reservoir populations is essential to predict and prevent future emerging diseases. Here, we identify a high incidence of multiple paramyxoviruses in urine samples collected from a closed captive colony of circa 115 straw-coloured fruit bats (Eidolon helvum). The sequences detected have high nucleotide identities with those derived from free ranging African fruit bats and form phylogenetic clusters with the Henipavirus genus, Pararubulavirus genus and other unclassified paramyxoviruses. As this colony had been closed for 5 years prior to this study, these results indicate that within-host paramyxoviral persistence underlies the role of bats as reservoirs of these viruses.

First Time Wheezing in a 2-Year-Old.

Pediatric patients with respiratory signs and symptoms who are found to be wheezing present a diagnostic dilemma to pediatricians. The majority of these cases are diagnosed as some degree of reactive airway disease, either as viral bronchiolitis or asthma. In this scenario, a patient with wheezing was initially given 2 courses of appropriate antibiotics on the basis of the duration and concurrence of other symptoms. However, he was subsequently referred to a pediatric pulmonologist for further workup after failure to improve and persistent oxygen saturations in the low-to-mid 90s. More extensive testing was completed by the pediatric pulmonologist, in addition to a short hospital admission. A rigid bronchoscopy was eventually completed, which revealed small pieces of partially digested material. Although his persistent cough resolved, his saturations continued to be suboptimal. A chest computed tomography scan with contrast was then completed, which eventually led to his diagnosis and appropriate treatment and resolution of his symptoms.

Prematurity, the diagnosis of bronchopulmonary dysplasia, and maturation of ventilatory control.

Infants born before 32 weeks gestational age and receiving respiratory support at 36 weeks postmenstrual age (PMA) are diagnosed with bronchopulmonary dysplasia (BPD). This label suggests that their need for supplemental oxygen (O2 ) is primarily due to acquired dysplasia of airways and airspaces, and that the supplemental O2 is treating residual parenchymal lung disease. However, emerging evidence suggests that immature ventilatory control may also contribute to the need for supplemental O2 at 36 weeks PMA. In all newborns, maturation of ventilatory control continues ex utero and is a plastic process. Among premature infants, supplemental O2 mitigates the hypoxemic effects of delayed maturation of ventilatory control, as well as reduces the duration and frequency of periodic breathing events. Nevertheless, prematurity is associated with altered and occasionally aberrant maturation of ventilatory control. Infants born prematurely, with or without a diagnosis of BPD, are more prone to long-lasting effects of dysfunctional ventilatory control. This review addresses normal and abnormal maturation of ventilatory control and suggests how aberrant maturation complicates assigning the diagnosis of BPD. Greater awareness of the interaction between parenchymal lung disease and delayed maturation of ventilatory control is essential to understanding why a given premature infant requires and is benefitting from supplemental O2 at 36 weeks PMA.

Superresolution light microscopy of the Drosophila histone locus body reveals a core-shell organization associated with expression of replication-dependent histone genes.

The histone locus body (HLB) is an evolutionarily conserved nuclear body that regulates the transcription and processing of replication-dependent (RD) histone mRNAs, which are the only eukaryotic mRNAs lacking a poly-A tail. Many nuclear bodies contain distinct domains, but how internal organization is related to nuclear body function is not fully understood. Here, we demonstrate using structured illumination microscopy that Drosophila HLBs have a "core-shell" organization in which the internal core contains transcriptionally active RD histone genes. The N-terminus of Mxc, which contains a domain required for Mxc oligomerization, HLB assembly, and RD histone gene expression, is enriched in the HLB core. In contrast, the C-terminus of Mxc is enriched in the HLB outer shell as is FLASH, a component of the active U7 snRNP that cotranscriptionally cleaves RD histone pre-mRNA. Consistent with these results, we show biochemically that FLASH binds directly to the Mxc C-terminal region. In the rapid S-M nuclear cycles of syncytial blastoderm Drosophila embryos, the HLB disassembles at mitosis and reassembles the core-shell arrangement as histone gene transcription is activated immediately after mitosis. Thus, the core-shell organization is coupled to zygotic histone gene transcription, revealing a link between HLB internal organization and RD histone gene expression.

CDK-Regulated Phase Separation Seeded by Histone Genes Ensures Precise Growth and Function of Histone Locus Bodies.

Many membraneless organelles form through liquid-liquid phase separation, but how their size is controlled and whether size is linked to function remain poorly understood. The histone locus body (HLB) is an evolutionarily conserved nuclear body that regulates the transcription and processing of histone mRNAs. Here, we show that Drosophila HLBs form through phase separation. During embryogenesis, the size of HLBs is controlled in a precise and dynamic manner that is dependent on the cell cycle and zygotic histone gene activation. Control of HLB growth is achieved by a mechanism integrating nascent mRNAs at the histone locus, which facilitates phase separation, and the nuclear concentration of the scaffold protein multi-sex combs (Mxc), which is controlled by the activity of cyclin-dependent kinases. Reduced Cdk2 activity results in smaller HLBs and the appearance of nascent, misprocessed histone mRNAs. Thus, our experiments identify a mechanism linking nuclear body growth and size with gene expression.

Benchmarking Deep Learning Architectures for Predicting Readmission to the ICU and Describing Patients-at-Risk.

To compare different deep learning architectures for predicting the risk of readmission within 30 days of discharge from the intensive care unit (ICU). The interpretability of attention-based models is leveraged to describe patients-at-risk. Several deep learning architectures making use of attention mechanisms, recurrent layers, neural ordinary differential equations (ODEs), and medical concept embeddings with time-aware attention were trained using publicly available electronic medical record data (MIMIC-III) associated with 45,298 ICU stays for 33,150 patients. Bayesian inference was used to compute the posterior over weights of an attention-based model. Odds ratios associated with an increased risk of readmission were computed for static variables. Diagnoses, procedures, medications, and vital signs were ranked according to the associated risk of readmission. A recurrent neural network, with time dynamics of code embeddings computed by neural ODEs, achieved the highest average precision of 0.331 (AUROC: 0.739, F1-Score: 0.372). Predictive accuracy was comparable across neural network architectures. Groups of patients at risk included those suffering from infectious complications, with chronic or progressive conditions, and for whom standard medical care was not suitable. Attention-based networks may be preferable to recurrent networks if an interpretable model is required, at only marginal cost in predictive accuracy.

Postmenstrual age at discharge in premature infants with and without ventilatory pattern instability.

RATIONALE: To determine if ventilatory pattern instability, manifested as periodic breathing (PB) during physiologic challenge testing, affects postmenstrual age (PMA) at discharge. METHODS: Eighty infants underwent challenge testing at 36 weeks PMA. Infants breathing supplemental O2 received a room air challenge (RAC, N = 51); those breathing ambient air underwent a hypoxic challenge test (HCT, N = 29). Infants were assigned one of four ventilatory control phenotypes based on the presence or absence of PB during their test, and if they passed or failed because of hypoxemia during the challenge test. RESULTS: There were no clinical or demographic differences between groups. Infants who passed their challenge testing were, on average, discharged 1.6 weeks sooner than those who failed. The groups of ventilatory control phenotypes differed in PMA at discharge (p = 0.0020), but those with PB were younger by PMA at discharge. CONCLUSIONS: Ventilatory pattern instability did not prolong time to discharge. Passing either challenge was associated with earlier discharge, suggesting these tests might identify infants who can have nasal cannula support removed and be safely discharged sooner. Most of the infants who failed their challenge tests with PB were receiving nasal cannula support. Nasal cannula support may be not only treating hypoxemia due to bronchopulmonary dysplasia (BPD), but also mitigating their ventilatory pattern instability.