Early prediction of hemodynamic interventions in the intensive care unit using machine learning.

BACKGROUND: Timely recognition of hemodynamic instability in critically ill patients enables increased vigilance and early treatment opportunities. We develop the Hemodynamic Stability Index (HSI), which highlights situational awareness of possible hemodynamic instability occurring at the bedside and to prompt assessment for potential hemodynamic interventions. METHODS: We used an ensemble of decision trees to obtain a real-time risk score that predicts the initiation of hemodynamic interventions an hour into the future. We developed the model using the eICU Research Institute (eRI) database, based on adult ICU admissions from 2012 to 2016. A total of 208,375 ICU stays met the inclusion criteria, with 32,896 patients (prevalence = 18%) experiencing at least one instability event where they received one of the interventions during their stay. Predictors included vital signs, laboratory measurements, and ventilation settings. RESULTS: HSI showed significantly better performance compared to single parameters like systolic blood pressure and shock index (heart rate/systolic blood pressure) and showed good generalization across patient subgroups. HSI AUC was 0.82 and predicted 52% of all hemodynamic interventions with a lead time of 1-h with a specificity of 92%. In addition to predicting future hemodynamic interventions, our model provides confidence intervals and a ranked list of clinical features that contribute to each prediction. Importantly, HSI can use a sparse set of physiologic variables and abstains from making a prediction when the confidence is below an acceptable threshold. CONCLUSIONS: The HSI algorithm provides a single score that summarizes hemodynamic status in real time using multiple physiologic parameters in patient monitors and electronic medical records (EMR). Importantly, HSI is designed for real-world deployment, demonstrating generalizability, strong performance under different data availability conditions, and providing model explanation in the form of feature importance and prediction confidence.

COPD subtypes identified by network-based clustering of blood gene expression.

One of the most common smoking-related diseases, chronic obstructive pulmonary disease (COPD), results from a dysregulated, multi-tissue inflammatory response to cigarette smoke. We hypothesized that systemic inflammatory signals in genome-wide blood gene expression can identify clinically important COPD-related disease subtypes, and we leveraged pre-existing gene interaction networks to guide unsupervised clustering of blood microarray expression data. Using network-informed non-negative matrix factorization, we analyzed genome-wide blood gene expression from 229 former smokers in the ECLIPSE Study, and we identified novel, clinically relevant molecular subtypes of COPD. These network-informed clusters were more stable and more strongly associated with measures of lung structure and function than clusters derived from a network-naïve approach, and they were associated with subtype-specific enrichment for inflammatory and protein catabolic pathways. These clusters were successfully reproduced in an independent sample of 135 smokers from the COPDGene Study.

Lidar measurements of solid rocket propellant fire particle plumes.

This paper presents the first, to our knowledge, direct measurement of aerosol produced by an aluminized solid rocket propellant (SRP) fire on the ground. Such fires produce aluminum oxide particles small enough to loft high into the atmosphere and disperse over a wide area. These results can be applied to spacecraft launchpad accidents that expose spacecraft to such fires; during these fires, there is concern that some of the plutonium from the spacecraft power system will be carried with the aerosols. Accident-related lofting of this material would be the net result of many contributing processes that are currently being evaluated. To resolve the complexity of fire processes, a self-consistent model of the ground-level and upper-level parts of the plume was determined by merging ground-level optical measurements of the fire with lidar measurements of the aerosol plume at height during a series of SRP fire tests that simulated propellant fire accident scenarios. On the basis of the measurements and model results, the Johns Hopkins University Applied Physics Laboratory (JHU/APL) team was able to estimate the amount of aluminum oxide (alumina) lofted into the atmosphere above the fire. The quantification of this ratio is critical for a complete understanding of accident scenarios, because contaminants are transported through the plume. This paper provides an estimate for the mass of alumina lofted into the air.

Cluster analysis in the COPDGene study identifies subtypes of smokers with distinct patterns of airway disease and emphysema.

BACKGROUND: There is notable heterogeneity in the clinical presentation of patients with COPD. To characterise this heterogeneity, we sought to identify subgroups of smokers by applying cluster analysis to data from the COPDGene study. METHODS: We applied a clustering method, k-means, to data from 10 192 smokers in the COPDGene study. After splitting the sample into a training and validation set, we evaluated three sets of input features across a range of k (user-specified number of clusters). Stable solutions were tested for association with four COPD-related measures and five genetic variants previously associated with COPD at genome-wide significance. The results were confirmed in the validation set. FINDINGS: We identified four clusters that can be characterised as (1) relatively resistant smokers (ie, no/mild obstruction and minimal emphysema despite heavy smoking), (2) mild upper zone emphysema-predominant, (3) airway disease-predominant and (4) severe emphysema. All clusters are strongly associated with COPD-related clinical characteristics, including exacerbations and dyspnoea (p<0.001). We found strong genetic associations between the mild upper zone emphysema group and rs1980057 near HHIP, and between the severe emphysema group and rs8034191 in the chromosome 15q region (p<0.001). All significant associations were replicated at p<0.05 in the validation sample (12/12 associations with clinical measures and 2/2 genetic associations). INTERPRETATION: Cluster analysis identifies four subgroups of smokers that show robust associations with clinical characteristics of COPD and known COPD-associated genetic variants.

Early Prediction of Cardiogenic Shock Using Machine Learning.

Cardiogenic shock (CS) is a severe condition with in-hospital mortality of up to 50%. Patients who develop CS may have previous cardiac history, but that may not always be the case, adding to the challenges in optimally identifying and managing these patients. Patients may present to a medical facility with CS or develop CS while in the emergency department (ED), in a general inpatient ward (WARD) or in the critical care unit (CC). While different clinical pathways for management exist once CS is recognized, there are challenges in identifying the patients in a timely manner, in all settings, in a timeframe that will allow proper management. We therefore developed and evaluated retrospectively a machine learning model based on the XGBoost (XGB) algorithm which runs automatically on patient data from the electronic health record (EHR). The algorithm was trained on 8 years of de-identified data (from 2010 to 2017) collected from a large regional healthcare system. The input variables include demographics, vital signs, laboratory values, some orders, and specific pre-existing diagnoses. The model was designed to make predictions 2 h prior to the need of first CS intervention (inotrope, vasopressor, or mechanical circulatory support). The algorithm achieves an overall area under curve (AUC) of 0.87 (0.81 in CC, 0.84 in ED, 0.97 in WARD), which is considered useful for clinical use. The algorithm can be refined based on specific elements defining patient subpopulations, for example presence of acute myocardial infarction (AMI) or congestive heart failure (CHF), further increasing its precision when a patient has these conditions. The top-contributing risk factors learned by the model are consistent with existing clinical findings. Our conclusion is that a useful machine learning model can be used to predict the development of CS. This manuscript describes the main steps of the development process and our results.

Machine Learning Prediction of Progression in Forced Expiratory Volume in 1 Second in the COPDGene® Study.

Background: The heterogeneous nature of chronic obstructive pulmonary disease (COPD) complicates the identification of the predictors of disease progression. We aimed to improve the prediction of disease progression in COPD by using machine learning and incorporating a rich dataset of phenotypic features. Methods: We included 4496 smokers with available data from their enrollment and 5-year follow-up visits in the COPD Genetic Epidemiology (COPDGene®) study. We constructed linear regression (LR) and supervised random forest models to predict 5-year progression in forced expiratory in 1 second (FEV1) from 46 baseline features. Using cross-validation, we randomly partitioned participants into training and testing samples. We also validated the results in the COPDGene 10-year follow-up visit. Results: Predicting the change in FEV1 over time is more challenging than simply predicting the future absolute FEV1 level. For random forest, R-squared was 0.15 and the area under the receiver operator characteristic (ROC) curves for the prediction of participants in the top quartile of observed progression was 0.71 (testing) and respectively, 0.10 and 0.70 (validation). Random forest provided slightly better performance than LR. The accuracy was best for Global initiative for chronic Obstructive Lung Disease (GOLD) grades 1-2 participants, and it was harder to achieve accurate prediction in advanced stages of the disease. Predictive variables differed in their relative importance as well as for the predictions by GOLD. Conclusion: Random forest, along with deep phenotyping, predicts FEV1 progression with reasonable accuracy. There is significant room for improvement in future models. This prediction model facilitates the identification of smokers at increased risk for rapid disease progression. Such findings may be useful in the selection of patient populations for targeted clinical trials.

Emotion fingerprints or emotion populations? A meta-analytic investigation of autonomic features of emotion categories.

The classical view of emotion hypothesizes that certain emotion categories have a specific autonomic nervous system (ANS) "fingerprint" that is distinct from other categories. Substantial ANS variation within a category is presumed to be epiphenomenal. The theory of constructed emotion hypothesizes that an emotion category is a population of context-specific, highly variable instances that need not share an ANS fingerprint. Instead, ANS variation within a category is a meaningful part of the nature of emotion. We present a meta-analysis of 202 studies measuring ANS reactivity during lab-based inductions of emotion in nonclinical samples of adults, using a random effects, multilevel meta-analysis and multivariate pattern classification analysis to test our hypotheses. We found increases in mean effect size for 59.4% of ANS variables across emotion categories, but the pattern of effect sizes did not clearly distinguish 1 emotion category from another. We also observed significant variation within emotion categories; heterogeneity accounted for a moderate to substantial percentage (i.e., I2 ≥ 30%) of variability in 54% of these effect sizes. Experimental moderators epiphenomenal to emotion, such as induction type (e.g., films vs. imagery), did not explain a large portion of the variability. Correction for publication bias reduced estimated effect sizes even further, increasing heterogeneity of effect sizes for certain emotion categories. These findings, when considered in the broader empirical literature, are more consistent with population thinking and other principles from evolutionary biology found within the theory of constructed emotion, and offer insights for developing new hypotheses to understand the nature of emotion. (PsycINFO Database Record

Lobar Emphysema Distribution Is Associated With 5-Year Radiological Disease Progression.

BACKGROUND: Emphysema has considerable variability in its regional distribution. Craniocaudal emphysema distribution is an important predictor of the response to lung volume reduction. However, there is little consensus regarding how to define upper lobe-predominant and lower lobe-predominant emphysema subtypes. Consequently, the clinical and genetic associations with these subtypes are poorly characterized. METHODS: We sought to identify subgroups characterized by upper-lobe or lower-lobe emphysema predominance and comparable amounts of total emphysema by analyzing data from 9,210 smokers without alpha-1-antitrypsin deficiency in the Genetic Epidemiology of COPD (COPDGene) cohort. CT densitometric emphysema was measured in each lung lobe. Random forest clustering was applied to lobar emphysema variables after regressing out the effects of total emphysema. Clusters were tested for association with clinical and imaging outcomes at baseline and at 5-year follow-up. Their associations with genetic variants were also compared. RESULTS: Three clusters were identified: minimal emphysema (n = 1,312), upper lobe-predominant emphysema (n = 905), and lower lobe-predominant emphysema (n = 796). Despite a similar amount of total emphysema, the lower-lobe group had more severe airflow obstruction at baseline and higher rates of metabolic syndrome compared with subjects with upper-lobe predominance. The group with upper-lobe predominance had greater 5-year progression of emphysema, gas trapping, and dyspnea. Differential associations with known COPD genetic risk variants were noted. CONCLUSIONS: Subgroups of smokers defined by upper-lobe or lower-lobe emphysema predominance exhibit different functional and radiological disease progression rates, and the upper-lobe predominant subtype shows evidence of association with known COPD genetic risk variants. These subgroups may be useful in the development of personalized treatments for COPD.

A Bayesian Nonparametric Model for Disease Subtyping: Application to Emphysema Phenotypes.

We introduce a novel Bayesian nonparametric model that uses the concept of disease trajectories for disease subtype identification. Although our model is general, we demonstrate that by treating fractions of tissue patterns derived from medical images as compositional data, our model can be applied to study distinct progression trends between population subgroups. Specifically, we apply our algorithm to quantitative emphysema measurements obtained from chest CT scans in the COPDGene Study and show several distinct progression patterns. As emphysema is one of the major components of chronic obstructive pulmonary disease (COPD), the third leading cause of death in the United States [1], an improved definition of emphysema and COPD subtypes is of great interest. We investigate several models with our algorithm, and show that one with age , pack years (a measure of cigarette exposure), and smoking status as predictors gives the best compromise between estimated predictive performance and model complexity. This model identified nine subtypes which showed significant associations to seven single nucleotide polymorphisms (SNPs) known to associate with COPD. Additionally, this model gives better predictive accuracy than multiple, multivariate ordinary least squares regression as demonstrated in a five-fold cross validation analysis. We view our subtyping algorithm as a contribution that can be applied to bridge the gap between CT-level assessment of tissue composition to population-level analysis of compositional trends that vary between disease subtypes.

Phenotypic and genetic heterogeneity among subjects with mild airflow obstruction in COPDGene.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is characterized by marked phenotypic heterogeneity. Most previous studies have focused on COPD subjects with FEV1 < 80% predicted. We investigated the clinical and genetic heterogeneity in subjects with mild airflow limitation in spirometry grade 1 defined by the Global Initiative for chronic Obstructive Lung Disease (GOLD 1). METHODS: Data from current and former smokers participating in the COPDGene Study (NCT00608764) were analyzed. K-means clustering was performed to explore subtypes within 794 GOLD 1 subjects. For all subjects with GOLD 1 and with each cluster, a genome-wide association study and candidate gene testing were performed using smokers with normal lung function as a control group. Combinations of COPD genome-wide significant single nucleotide polymorphisms (SNPs) were tested for association with FEV1 (% predicted) in GOLD 1 and in a combined group of GOLD 1 and smoking control subjects. RESULTS: K-means clustering of GOLD 1 subjects identified putative "near-normal", "airway-predominant", "emphysema-predominant" and "lowest FEV1% predicted" subtypes. In non-Hispanic whites, the only SNP nominally associated with GOLD 1 status relative to smoking controls was rs7671167 (FAM13A) in logistic regression models with adjustment for age, sex, pack-years of smoking, and genetic ancestry. The emphysema-predominant GOLD 1 cluster was nominally associated with rs7671167 (FAM13A) and rs161976 (BICD1). The lowest FEV1% predicted cluster was nominally associated with rs1980057 (HHIP) and rs1051730 (CHRNA3). Combinations of COPD genome-wide significant SNPs were associated with FEV1 (% predicted) in a combined group of GOLD 1 and smoking control subjects. CONCLUSIONS: Our results indicate that GOLD 1 subjects show substantial clinical heterogeneity, which is at least partially related to genetic heterogeneity.