Quantifiable identification of flow-limited ventilator dyssynchrony with the deformed lung ventilator model.

BACKGROUND: Ventilator dyssynchrony (VD) can worsen lung injury and is challenging to detect and quantify due to the complex variability in the dyssynchronous breaths. While machine learning (ML) approaches are useful for automating VD detection from the ventilator waveform data, scalable severity quantification and its association with pathogenesis and ventilator mechanics remain challenging. OBJECTIVE: We develop a systematic framework to quantify pathophysiological features observed in ventilator waveform signals such that they can be used to create feature-based severity stratification of VD breaths. METHODS: A mathematical model was developed to represent the pressure and volume waveforms of individual breaths in a feature-based parametric form. Model estimates of respiratory effort strength were used to assess the severity of flow-limited (FL)-VD breaths compared to normal breaths. A total of 93,007 breath waveforms from 13 patients were analyzed. RESULTS: A novel model-defined continuous severity marker was developed and used to estimate breath phenotypes of FL-VD breaths. The phenotypes had a predictive accuracy of over 97% with respect to the previously developed ML-VD identification algorithm. To understand the incidence of FL-VD breaths and their association with the patient state, these phenotypes were further successfully correlated with ventilator-measured parameters and electronic health records. CONCLUSION: This work provides a computational pipeline to identify and quantify the severity of FL-VD breaths and paves the way for a large-scale study of VD causes and effects. This approach has direct application to clinical practice and in meaningful knowledge extraction from the ventilator waveform data.

Development and Validation of the Phoenix Criteria for Pediatric Sepsis and Septic Shock.

Importance: The Society of Critical Care Medicine Pediatric Sepsis Definition Task Force sought to develop and validate new clinical criteria for pediatric sepsis and septic shock using measures of organ dysfunction through a data-driven approach. Objective: To derive and validate novel criteria for pediatric sepsis and septic shock across differently resourced settings. Design, Setting, and Participants: Multicenter, international, retrospective cohort study in 10 health systems in the US, Colombia, Bangladesh, China, and Kenya, 3 of which were used as external validation sites. Data were collected from emergency and inpatient encounters for children (aged <18 years) from 2010 to 2019: 3 049 699 in the development (including derivation and internal validation) set and 581 317 in the external validation set. Exposure: Stacked regression models to predict mortality in children with suspected infection were derived and validated using the best-performing organ dysfunction subscores from 8 existing scores. The final model was then translated into an integer-based score used to establish binary criteria for sepsis and septic shock. Main Outcomes and Measures: The primary outcome for all analyses was in-hospital mortality. Model- and integer-based score performance measures included the area under the precision recall curve (AUPRC; primary) and area under the receiver operating characteristic curve (AUROC; secondary). For binary criteria, primary performance measures were positive predictive value and sensitivity. Results: Among the 172 984 children with suspected infection in the first 24 hours (development set; 1.2% mortality), a 4-organ-system model performed best. The integer version of that model, the Phoenix Sepsis Score, had AUPRCs of 0.23 to 0.38 (95% CI range, 0.20-0.39) and AUROCs of 0.71 to 0.92 (95% CI range, 0.70-0.92) to predict mortality in the validation sets. Using a Phoenix Sepsis Score of 2 points or higher in children with suspected infection as criteria for sepsis and sepsis plus 1 or more cardiovascular point as criteria for septic shock resulted in a higher positive predictive value and higher or similar sensitivity compared with the 2005 International Pediatric Sepsis Consensus Conference (IPSCC) criteria across differently resourced settings. Conclusions and Relevance: The novel Phoenix sepsis criteria, which were derived and validated using data from higher- and lower-resource settings, had improved performance for the diagnosis of pediatric sepsis and septic shock compared with the existing IPSCC criteria.

Intracranial pressure-flow relationships in traumatic brain injury patients expose gaps in the tenets of models and pressure-oriented management.

Background: The protocols and therapeutic guidance established for treating traumatic brain injuries (TBI) in neurointensive care focus on managing cerebral blood flow (CBF) and brain tissue oxygenation based on pressure signals. The decision support process relies on assumed relationships between cerebral perfusion pressure (CPP) and blood flow, pressure-flow relationships (PFRs), and shares this framework of assumptions with mathematical intracranial hemodynamic models. These foundational assumptions are difficult to verify, and their violation can impact clinical decision-making and model validity. Method: A hypothesis- and model-driven method for verifying and understanding the foundational intracranial hemodynamic PFRs is developed and applied to a novel multi-modality monitoring dataset. Results: Model analysis of joint observations of CPP and CBF validates the standard PFR when autoregulatory processes are impaired as well as unmodelable cases dominated by autoregulation. However, it also identifies a dynamical regime -or behavior pattern- where the PFR assumptions are wrong in a precise, data-inferable way due to negative CPP-CBF coordination over long timescales. This regime is of both clinical and research interest: its dynamics are modelable under modified assumptions while its causal direction and mechanistic pathway remain unclear. Conclusions: Motivated by the understanding of mathematical physiology, the validity of the standard PFR can be assessed a) directly by analyzing pressure reactivity and mean flow indices (PRx and Mx) or b) indirectly through the relationship between CBF and other clinical observables. This approach could potentially help personalize TBI care by considering intracranial pressure and CPP in relation to other data, particularly CBF. The analysis suggests a threshold using clinical indices of autoregulation jointly generalizes independently set indicators to assess CA functionality. These results support the use of increasingly data-rich environments to develop more robust hybrid physiological-machine learning models.

Flow-Limited and Reverse-Triggered Ventilator Dyssynchrony Are Associated With Increased Tidal and Dynamic Transpulmonary Pressure.

OBJECTIVES: Ventilator dyssynchrony may be associated with increased delivered tidal volumes (V t s) and dynamic transpulmonary pressure (ΔP L,dyn ), surrogate markers of lung stress and strain, despite low V t ventilation. However, it is unknown which types of ventilator dyssynchrony are most likely to increase these metrics or if specific ventilation or sedation strategies can mitigate this potential. DESIGN: A prospective cohort analysis to delineate the association between ten types of breaths and delivered V t , ΔP L,dyn , and transpulmonary mechanical energy. SETTING: Patients admitted to the medical ICU. PATIENTS: Over 580,000 breaths from 35 patients with acute respiratory distress syndrome (ARDS) or ARDS risk factors. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Patients received continuous esophageal manometry. Ventilator dyssynchrony was identified using a machine learning algorithm. Mixed-effect models predicted V t , ΔP L,dyn , and transpulmonary mechanical energy for each type of ventilator dyssynchrony while controlling for repeated measures. Finally, we described how V t , positive end-expiratory pressure (PEEP), and sedation (Richmond Agitation-Sedation Scale) strategies modify ventilator dyssynchrony's association with these surrogate markers of lung stress and strain. Double-triggered breaths were associated with the most significant increase in V t , ΔP L,dyn , and transpulmonary mechanical energy. However, flow-limited, early reverse-triggered, and early ventilator-terminated breaths were also associated with significant increases in V t , ΔP L,dyn , and energy. The potential of a ventilator dyssynchrony type to increase V t , ΔP L,dyn , or energy clustered similarly. Increasing set V t may be associated with a disproportionate increase in high-volume and high-energy ventilation from double-triggered breaths, but PEEP and sedation do not clinically modify the interaction between ventilator dyssynchrony and surrogate markers of lung stress and strain. CONCLUSIONS: Double-triggered, flow-limited, early reverse-triggered, and early ventilator-terminated breaths are associated with increases in V t , ΔP L,dyn , and energy. As flow-limited breaths are more than twice as common as double-triggered breaths, further work is needed to determine the interaction of ventilator dyssynchrony frequency to cause clinically meaningful changes in patient outcomes.

International Consensus Criteria for Pediatric Sepsis and Septic Shock.

Importance: Sepsis is a leading cause of death among children worldwide. Current pediatric-specific criteria for sepsis were published in 2005 based on expert opinion. In 2016, the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) defined sepsis as life-threatening organ dysfunction caused by a dysregulated host response to infection, but it excluded children. Objective: To update and evaluate criteria for sepsis and septic shock in children. Evidence Review: The Society of Critical Care Medicine (SCCM) convened a task force of 35 pediatric experts in critical care, emergency medicine, infectious diseases, general pediatrics, nursing, public health, and neonatology from 6 continents. Using evidence from an international survey, systematic review and meta-analysis, and a new organ dysfunction score developed based on more than 3 million electronic health record encounters from 10 sites on 4 continents, a modified Delphi consensus process was employed to develop criteria. Findings: Based on survey data, most pediatric clinicians used sepsis to refer to infection with life-threatening organ dysfunction, which differed from prior pediatric sepsis criteria that used systemic inflammatory response syndrome (SIRS) criteria, which have poor predictive properties, and included the redundant term, severe sepsis. The SCCM task force recommends that sepsis in children be identified by a Phoenix Sepsis Score of at least 2 points in children with suspected infection, which indicates potentially life-threatening dysfunction of the respiratory, cardiovascular, coagulation, and/or neurological systems. Children with a Phoenix Sepsis Score of at least 2 points had in-hospital mortality of 7.1% in higher-resource settings and 28.5% in lower-resource settings, more than 8 times that of children with suspected infection not meeting these criteria. Mortality was higher in children who had organ dysfunction in at least 1 of 4-respiratory, cardiovascular, coagulation, and/or neurological-organ systems that was not the primary site of infection. Septic shock was defined as children with sepsis who had cardiovascular dysfunction, indicated by at least 1 cardiovascular point in the Phoenix Sepsis Score, which included severe hypotension for age, blood lactate exceeding 5 mmol/L, or need for vasoactive medication. Children with septic shock had an in-hospital mortality rate of 10.8% and 33.5% in higher- and lower-resource settings, respectively. Conclusions and Relevance: The Phoenix sepsis criteria for sepsis and septic shock in children were derived and validated by the international SCCM Pediatric Sepsis Definition Task Force using a large international database and survey, systematic review and meta-analysis, and modified Delphi consensus approach. A Phoenix Sepsis Score of at least 2 identified potentially life-threatening organ dysfunction in children younger than 18 years with infection, and its use has the potential to improve clinical care, epidemiological assessment, and research in pediatric sepsis and septic shock around the world.

The Development, Optimization, and Validation of Four Different Machine Learning Algorithms to Identify Ventilator Dyssynchrony.

Invasive mechanical ventilation can worsen lung injury. Ventilator dyssynchrony (VD) may propagate ventilator-induced lung injury (VILI) and is challenging to detect and systematically monitor because each patient takes approximately 25,000 breaths a day yet some types of VD are rare, accounting for less than 1% of all breaths. Therefore, we sought to develop and validate accurate machine learning (ML) algorithms to detect multiple types of VD by leveraging esophageal pressure waveform data to quantify patient effort with airway pressure, flow, and volume data generated during mechanical ventilation, building a computational pipeline to facilitate the study of VD. Materials and Methods: We collected ventilator waveform and esophageal pressure data from 30 patients admitted to the ICU. Esophageal pressure allows the measurement of transpulmonary pressure and patient effort. Waveform data were cleaned, features considered essential to VD detection were calculated, and a set of 10,000 breaths were manually labeled. Four ML algorithms were trained to classify each type of VD: logistic regression, support vector classification, random forest, and XGBoost. Results: We trained ML models to detect different families and seven types of VD with high sensitivity (>90% and >80%, respectively). Three types of VD remained difficult for ML to classify because of their rarity and lack of sample size. XGBoost classified breaths with increased specificity compared to other ML algorithms. Discussion: We developed ML models to detect multiple types of VD accurately. The ability to accurately detect multiple VD types addresses one of the significant limitations in understanding the role of VD in affecting patient outcomes. Conclusion: ML models identify multiple types of VD by utilizing esophageal pressure data and airway pressure, flow, and volume waveforms. The development of such computational pipelines will facilitate the identification of VD in a scalable fashion, allowing for the systematic study of VD and its impact on patient outcomes.

ß-cell intrinsic dynamics rather than gap junction structure dictates subpopulations in the islet functional network.

Diabetes is caused by the inability of electrically coupled, functionally heterogeneous ß-cells within the pancreatic islet to provide adequate insulin secretion. Functional networks have been used to represent synchronized oscillatory [Ca2+] dynamics and to study ß-cell subpopulations, which play an important role in driving islet function. The mechanism by which highly synchronized ß-cell subpopulations drive islet function is unclear. We used experimental and computational techniques to investigate the relationship between functional networks, structural (gap junction) networks, and intrinsic ß-cell dynamics in slow and fast oscillating islets. Highly synchronized subpopulations in the functional network were differentiated by intrinsic dynamics, including metabolic activity and KATP channel conductance, more than structural coupling. Consistent with this, intrinsic dynamics were more predictive of high synchronization in the islet functional network as compared to high levels of structural coupling. Finally, dysfunction of gap junctions, which can occur in diabetes, caused decreases in the efficiency and clustering of the functional network. These results indicate that intrinsic dynamics rather than structure drive connections in the functional network and highly synchronized subpopulations, but gap junctions are still essential for overall network efficiency. These findings deepen our interpretation of functional networks and the formation of functional subpopulations in dynamic tissues such as the islet.

Differential effects of two-hit models of acute and ventilator-induced lung injury on lung structure, function, and inflammation.

Acute respiratory distress syndrome (ARDS) and acute lung injury have a diverse spectrum of causative factors including sepsis, aspiration of gastric contents, and near drowning. Clinical management of severe lung injury typically includes mechanical ventilation to maintain gas exchange which can lead to ventilator-induced lung injury (VILI). The cause of respiratory failure is acknowledged to affect the degree of lung inflammation, changes in lung structure, and the mechanical function of the injured lung. However, these differential effects of injury and the role of etiology in the structure-function relationship are not fully understood. To address this knowledge gap we caused lung injury with intratracheal hydrochloric acid (HCL) or endotoxin (LPS) 2 days prior to ventilation or with an injurious lavage (LAV) immediately prior to ventilation. These injury groups were then ventilated with high inspiratory pressures and positive end expiratory pressure (PEEP) = 0 cmH2O to cause VILI and model the clinical course of ARDS followed by supportive ventilation. The effects of injury were quantified using invasive lung function measurements recorded during PEEP ladders where the end-expiratory pressure was increased from 0 to 15 cm H2O and decreased back to 0 cmH2O in steps of 3 cmH2O. Design-based stereology was used to quantify the parenchymal structure of lungs air-inflated to 2, 5, and 10 cmH2O. Pro-inflammatory gene expression was measured with real-time quantitative polymerase chain reaction and alveolocapillary leak was estimated by measuring bronchoalveolar lavage protein content. The LAV group had small, stiff lungs that were recruitable at higher pressures, but did not demonstrate substantial inflammation. The LPS group showed septal swelling and high pro-inflammatory gene expression that was exacerbated by VILI. Despite widespread alveolar collapse, elastance in LPS was only modestly elevated above healthy mice (CTL) and there was no evidence of recruitability. The HCL group showed increased elastance and some recruitability, although to a lesser degree than LAV. Pro-inflammatory gene expression was elevated, but less than LPS, and the airspace dimensions were reduced. Taken together, those data highlight how different modes of injury, in combination with a 2nd hit of VILI, yield markedly different effects.

A simple modeling framework for prediction in the human glucose-insulin system.

Forecasting blood glucose (BG) levels with routinely collected data is useful for glycemic management. BG dynamics are nonlinear, complex, and nonstationary, which can be represented by nonlinear models. However, the sparsity of routinely collected data creates parameter identifiability issues when high-fidelity complex models are used, thereby resulting in inaccurate forecasts. One can use models with reduced physiological fidelity for robust and accurate parameter estimation and forecasting with sparse data. For this purpose, we approximate the nonlinear dynamics of BG regulation by a linear stochastic differential equation: we develop a linear stochastic model, which can be specialized to different settings: type 2 diabetes mellitus (T2DM) and intensive care unit (ICU), with different choices of appropriate model functions. The model includes deterministic terms quantifying glucose removal from the bloodstream through the glycemic regulation system and representing the effect of nutrition and externally delivered insulin. The stochastic term encapsulates the BG oscillations. The model output is in the form of an expected value accompanied by a band around this value. The model parameters are estimated patient-specifically, leading to personalized models. The forecasts consist of values for BG mean and variation, quantifying possible high and low BG levels. Such predictions have potential use for glycemic management as part of control systems. We present experimental results on parameter estimation and forecasting in T2DM and ICU settings. We compare the model's predictive capability with two different nonlinear models built for T2DM and ICU contexts to have a sense of the level of prediction achieved by this model.

Who needs what (features) when? Personalizing engagement with data-driven self-management to improve health equity.

OBJECTIVES: To examine the feasibility of promoting engagement with data-driven self-management of health among individuals from minoritized medically underserved communities by tailoring the design of self-management interventions to individuals' type of motivation and regulation in accordance with the Self-Determination Theory. METHODS: Fifty-three individuals with type 2 diabetes from an impoverished minority community were randomly assigned to four different versions of an mHealth app for data-driven self-management with the focus on nutrition, Platano; each version was tailored to a specific type of motivation and regulation within the SDT self-determination continuum. These versions included financial rewards (external regulation), feedback from expert registered dietitians (RDF, introjected regulation), self-assessment of attainment of one's nutritional goals (SA, identified regulation), and personalized meal-time nutrition decision support with post-meal blood glucose forecasts (FORC, integrated regulation). We used qualitative interviews to examine interaction between participants' experiences with the app and their motivation type (internal-external). RESULTS: As hypothesized, we found a clear interaction between the type of motivation and Platano features that users responded to and benefited from. For example, those with more internal motivation reported more positive experience with SA and FORC than those with more external motivation. However, we also found that Platano features that aimed to specifically address the needs of individuals with external regulation did not create the desired experience. We attribute this to a mismatch in emphasis on informational versus emotional support, particularly evident in RDF. In addition, we found that for participants recruited from an economically disadvantaged community, internal factors, such as motivation and regulation, interacted with external factors, most notably with limited health literacy and limited access to resources. CONCLUSIONS: The study suggests feasibility of using SDT to tailor design of mHealth interventions for promoting data-driven self-management to individuals' motivation and regulation. However, further research is needed to better align design solutions with different levels of self-determination continuum, to incorporate stronger emphasis on emotional support for individuals with external regulation, and to address unique needs and challenges of underserved communities, with particular attention to limited health literacy and access to resources.

Hypothesis-driven modeling of the human lung-ventilator system: A characterization tool for Acute Respiratory Distress Syndrome research.

Mechanical ventilation is an essential tool in the management of Acute Respiratory Distress Syndrome (ARDS), but it exposes patients to the risk of ventilator-induced lung injury (VILI). The human lung-ventilator system (LVS) involves the interaction of complex anatomy with a mechanical apparatus, which limits the ability of process-based models to provide individualized clinical support. This work proposes a hypothesis-driven strategy for LVS modeling in which robust personalization is achieved using a pre-defined parameter basis in a non-physiological model. Model inversion, here via windowed data assimilation, forges observed waveforms into interpretable parameter values that characterize the data rather than quantifying physiological processes. Accurate, model-based inference on human-ventilator data indicates model flexibility and utility over a variety of breath types, including those from dyssynchronous LVSs. Estimated parameters generate static characterizations of the data that are 50%-70% more accurate than breath-wise single-compartment model estimates. They also retain sufficient information to distinguish between the types of breath they represent. However, the fidelity and interpretability of model characterizations are tied to parameter definitions and model resolution. These additional factors must be considered in conjunction with the objectives of specific applications, such as identifying and tracking the development of human VILI.

Identifying low-dimensional trajectories of mechanically-ventilated patient systems: Empirical phenotypes of joint patient+care processes to enhance temporal analysis in ARDS research.

Mechanically ventilated patients generate waveform data that corresponds to patient interaction with unnatural forcing. This breath information includes both patient and apparatus sources, imbuing data with broad heterogeneity resulting from ventilator settings, patient efforts, patient-ventilator dyssynchronies, injuries, and other clinical therapies. Lung-protective ventilator settings outlined in respiratory care protocols lack personalization, and the connections between clinical outcomes and injuries resulting from mechanical ventilation remain poorly understood. Intra- and inter-patient heterogeneity and the volume of data comprising lung-ventilator system (LVS) observations limit broader and longer-time analysis of such systems. This work presents a computational pipeline for resolving LVS systems by tracking the evolution of data-conditioned model parameters and ventilator information. For individuals, the method presents LVS trajectory in a manageable way through low-dimensional representation of phenotypic breath waveforms. More general phenotypes across patients are also developed by aggregating patient-personalized estimates with additional normalization. The effectiveness of this process is demonstrated through application to multi-day observational series of 35 patients, which reveals the complexity of changes in the LVS over time. Considerable variations in breath behavior independent of the ventilator are revealed, suggesting the need to incorporate care factors such as patient sedation and posture in future analysis. The pipeline also identifies structural similarity in pressure-volume (pV) loop characterizations at the cohort level. The design invites active learning to incorporate clinical practitioner expertise into various methodological stages and algorithm choices.

Data assimilation on mechanistic models of glucose metabolism predicts glycemic states in adolescents following bariatric surgery.

Type 2 diabetes mellitus is a complex and under-treated disorder closely intertwined with obesity. Adolescents with severe obesity and type 2 diabetes have a more aggressive disease compared to adults, with a rapid decline in pancreatic ß cell function and increased incidence of comorbidities. Given the relative paucity of pharmacotherapies, bariatric surgery has become increasingly used as a therapeutic option. However, subsets of this population have sub-optimal outcomes with either inadequate weight loss or little improvement in disease. Predicting which patients will benefit from surgery is a difficult task and detailed physiological characteristics of patients who do not respond to treatment are generally unknown. Identifying physiological predictors of surgical response therefore has the potential to reveal both novel phenotypes of disease as well as therapeutic targets. We leverage data assimilation paired with mechanistic models of glucose metabolism to estimate pre-operative physiological states of bariatric surgery patients, thereby identifying latent phenotypes of impaired glucose metabolism. Specifically, maximal insulin secretion capacity, σ, and insulin sensitivity, SI, differentiate aberrations in glucose metabolism underlying an individual's disease. Using multivariable logistic regression, we combine clinical data with data assimilation to predict post-operative glycemic outcomes at 12 months. Models using data assimilation sans insulin had comparable performance to models using oral glucose tolerance test glucose and insulin. Our best performing models used data assimilation and had an area under the receiver operating characteristic curve of 0.77 (95% confidence interval 0.7665, 0.7734) and mean average precision of 0.6258 (0.6206, 0.6311). We show that data assimilation extracts knowledge from mechanistic models of glucose metabolism to infer future glycemic states from limited clinical data. This method can provide a pathway to predict long-term, post-surgical glycemic states by estimating the contributions of insulin resistance and limitations of insulin secretion to pre-operative glucose metabolism.

Inferring Insulin Secretion Rate from Sparse Patient Glucose and Insulin Measures.

The insulin secretion rate (ISR) contains information that can provide a personal, quantitative understanding of endocrine function. If the ISR can be reliably inferred from measurements, it could be used for understanding and clinically diagnosing problems with the glucose regulation system. Objective: This study aims to develop a model-based method for inferring a parametrization of the ISR and related physiological information among people with different glycemic conditions in a robust manner. The developed algorithm is applicable for both dense or sparsely sampled plasma glucose/insulin measurements, where sparseness is defined in terms of sampling time with respect to the fastest time scale of the dynamics. Methods: An algorithm for parametrizing and validating a functional form of the ISR for different compartmental models with unknown but estimable ISR function and absorption/decay rates describing the dynamics of insulin accumulation was developed. The method and modeling applies equally to c-peptide secretion rate (CSR) when c-peptide is measured. Accuracy of fit is reliant on reconstruction error of the measured trajectories, and when c-peptide is measured the relationship between CSR and ISR. The algorithm was applied to data from 17 subjects with normal glucose regulatory systems and 9 subjects with cystic fibrosis related diabetes (CFRD) in which glucose, insulin and c-peptide were measured in course of oral glucose tolerance tests (OGTT). Results: This model-based algorithm inferred parametrization of the ISR and CSR functional with relatively low reconstruction error for 12 of 17 control and 7 of 9 CFRD subjects. We demonstrate that when there are suspect measurements points, the validity of excluding them may be interrogated with this method. Significance: A new estimation method is available to infer the ISR and CSR functional profile along with plasma insulin and c-peptide absorption rates from sparse measurements of insulin, c-peptide, and plasma glucose concentrations. We propose a method to interrogate and exclude potentially erroneous OGTT measurement points based on reconstruction errors.

Evaluating Prediction of Continuous Clinical Values: A Glucose Case Study.

BACKGROUND: It would be useful to be able to assess the utility of predictive models of continuous values before clinical trials are performed. OBJECTIVE: The aim of the study is to compare metrics to assess the potential clinical utility of models that produce continuous value forecasts. METHODS: We ran a set of data assimilation forecast algorithms on time series of glucose measurements from neurological intensive care unit patients. We evaluated the forecasts using four sets of metrics: glucose root mean square (RMS) error, a set of metrics on a transformed glucose value, the estimated effect on clinical care based on an insulin guideline, and a glucose measurement error grid (Parkes grid). We assessed correlation among the metrics and created a set of factor models. RESULTS: The metrics generally correlated with each other, but those that estimated the effect on clinical care correlated with others the least and were generally associated with their own independent factors. The other metrics appeared to separate into those that emphasized errors in low glucose versus errors in high glucose. The Parkes grid was well correlated with the transformed glucose but not the estimation of clinical care. DISCUSSION: Our results indicate that we need to be careful before we assume that commonly used metrics like RMS error in raw glucose or even metrics like the Parkes grid that are designed to measure importance of differences will correlate well with actual effect on clinical care processes. A combination of metrics appeared to explain the most variance between cases. As prediction algorithms move into practice, it will be important to measure actual effects.

A Damaged-Informed Lung Ventilator Model for Ventilator Waveforms.

Motivated by a desire to understand pulmonary physiology, scientists have developed physiological lung models of varying complexity. However, pathophysiology and interactions between human lungs and ventilators, e.g., ventilator-induced lung injury (VILI), present challenges for modeling efforts. This is because the real-world pressure and volume signals may be too complex for simple models to capture, and while complex models tend not to be estimable with clinical data, limiting clinical utility. To address this gap, in this manuscript we developed a new damaged-informed lung ventilator (DILV) model. This approach relies on mathematizing ventilator pressure and volume waveforms, including lung physiology, mechanical ventilation, and their interaction. The model begins with nominal waveforms and adds limited, clinically relevant, hypothesis-driven features to the waveform corresponding to pulmonary pathophysiology, patient-ventilator interaction, and ventilator settings. The DILV model parameters uniquely and reliably recapitulate these features while having enough flexibility to reproduce commonly observed variability in clinical (human) and laboratory (mouse) waveform data. We evaluate the proof-in-principle capabilities of our modeling approach by estimating 399 breaths collected for differently damaged lungs for tightly controlled measurements in mice and uncontrolled human intensive care unit data in the absence and presence of ventilator dyssynchrony. The cumulative value of mean squares error for the DILV model is, on average, ≈12 times less than the single compartment lung model for all the waveforms considered. Moreover, changes in the estimated parameters correctly correlate with known measures of lung physiology, including lung compliance as a baseline evaluation. Our long-term goal is to use the DILV model for clinical monitoring and research studies by providing high fidelity estimates of lung state and sources of VILI with an end goal of improving management of VILI and acute respiratory distress syndrome.

Correction: Personalized glucose forecasting for type 2 diabetes using data assimilation.

[This corrects the article DOI: 10.1371/journal.pcbi.1005232.].

From Reflection to Action: Combining Machine Learning with Expert Knowledge for Nutrition Goal Recommendations.

Self-tracking can help personalize self-management interventions for chronic conditions like type 2 diabetes (T2D), but reflecting on personal data requires motivation and literacy. Machine learning (ML) methods can identify patterns, but a key challenge is making actionable suggestions based on personal health data. We introduce GlucoGoalie, which combines ML with an expert system to translate ML output into personalized nutrition goal suggestions for individuals with T2D. In a controlled experiment, participants with T2D found that goal suggestions were understandable and actionable. A 4-week in-the-wild deployment study showed that receiving goal suggestions augmented participants' self-discovery, choosing goals highlighted the multifaceted nature of personal preferences, and the experience of following goals demonstrated the importance of feedback and context. However, we identified tensions between abstract goals and concrete eating experiences and found static text too ambiguous for complex concepts. We discuss implications for ML-based interventions and the need for systems that offer more interactivity, feedback, and negotiation.