Inductive database to support iterative data mining: Application to biomarker analysis on patient data in the Fight-HF project.

Machine learning is now an essential part of any biomedical study but its integration into real effective Learning Health Systems, including the whole process of Knowledge Discovery from Data (KDD), is not yet realised. We propose an original extension of the KDD process model that involves an inductive database. We designed for the first time a generic model of Inductive Clinical DataBase (ICDB) aimed at hosting both patient data and learned models. We report experiments conducted on patient data in the frame of a project dedicated to fight heart failure. The results show how the ICDB approach allows to identify biomarker combinations, specific and predictive of heart fibrosis phenotype, that put forward hypotheses relative to underlying mechanisms. Two main scenarios were considered, a local-to-global KDD scenario and a trans-cohort alignment scenario. This promising proof of concept enables us to draw the contours of a next-generation Knowledge Discovery Environment (KDE).

Correction: Diagnostic performance of congestion score index evaluated from chest radiography for acute heart failure in the emergency department: A retrospective analysis from the PARADISE cohort.

[This corrects the article DOI: 10.1371/journal.pmed.1003419.].

Diagnostic performance of congestion score index evaluated from chest radiography for acute heart failure in the emergency department: A retrospective analysis from the PARADISE cohort.

BACKGROUND: Congestion score index (CSI), a semiquantitative evaluation of congestion on chest radiography (CXR), is associated with outcome in patients with heart failure (HF). However, its diagnostic value in patients admitted for acute dyspnea has yet to be evaluated. METHODS AND FINDINGS: The diagnostic value of CSI for acute HF (AHF; adjudicated from patients' discharge files) was studied in the Pathway of dyspneic patients in Emergency (PARADISE) cohort, including patients aged 18 years or older admitted for acute dyspnea in the emergency department (ED) of the Nancy University Hospital (France) between January 1, 2015 and December 31, 2015. CSI (ranging from 0 to 3) was evaluated using a semiquantitative method on CXR in consecutive patients admitted for acute dyspnea in the ED. Results were validated in independent cohorts (N = 224). Of 1,333 patients, mean (standard deviation [SD]) age was 72.0 (18.5) years, 686 (51.5%) were men, and mean (SD) CSI was 1.42 (0.79). Patients with higher CSI had more cardiovascular comorbidities, more severe congestion, higher b-type natriuretic peptide (BNP), poorer renal function, and more respiratory acidosis. AHF was diagnosed in 289 (21.7%) patients. CSI was significantly associated with AHF diagnosis (adjusted odds ratio [OR] for 0.1 unit CSI increase 1.19, 95% CI 1.16-1.22, p < 0.001) after adjustment for clinical-based diagnostic score including age, comorbidity burden, dyspnea, and clinical congestion. The diagnostic accuracy of CSI for AHF was >0.80, whether alone (area under the receiver operating characteristic curve [AUROC] 0.84, 95% CI 0.82-0.86) or in addition to the clinical model (AUROC 0.87, 95% CI 0.85-0.90). CSI improved diagnostic accuracy on top of clinical variables (net reclassification improvement [NRI] = 94.9%) and clinical variables plus BNP (NRI = 55.0%). Similar diagnostic accuracy was observed in the validation cohorts (AUROC 0.75, 95% CI 0.68-0.82). The key limitation of our derivation cohort was its single-center and retrospective nature, which was counterbalanced by the validation in the independent cohorts. CONCLUSIONS: In this study, we observed that a systematic semiquantified assessment of radiographic pulmonary congestion showed high diagnostic value for AHF in dyspneic patients. Better use of CXR may provide an inexpensive, widely, and readily available method for AHF triage in the ED.

Enhanced clinical phenotyping by mechanistic bioprofiling in heart failure with preserved ejection fraction: insights from the MEDIA-DHF study (The Metabolic Road to Diastolic Heart Failure).

Background: Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous syndrome for which clear evidence of effective therapies is lacking. Understanding which factors determine this heterogeneity may be helped by better phenotyping. An unsupervised statistical approach applied to a large set of biomarkers may identify distinct HFpEF phenotypes.Methods: Relevant proteomic biomarkers were analyzed in 392 HFpEF patients included in Metabolic Road to Diastolic HF (MEDIA-DHF). We performed an unsupervised cluster analysis to define distinct phenotypes. Cluster characteristics were explored with logistic regression. The association between clusters and 1-year cardiovascular (CV) death and/or CV hospitalization was studied using Cox regression.Results: Based on 415 biomarkers, we identified 2 distinct clusters. Clinical variables associated with cluster 2 were diabetes, impaired renal function, loop diuretics and/or betablockers. In addition, 17 biomarkers were higher expressed in cluster 2 vs. 1. Patients in cluster 2 vs. those in 1 experienced higher rates of CV death/CV hospitalization (adj. HR 1.93, 95% CI 1.12-3.32, p = 0.017). Complex-network analyses linked these biomarkers to immune system activation, signal transduction cascades, cell interactions and metabolism.Conclusion: Unsupervised machine-learning algorithms applied to a wide range of biomarkers identified 2 HFpEF clusters with different CV phenotypes and outcomes. The identified pathways may provide a basis for future research.Clinical significanceMore insight is obtained in the mechanisms related to poor outcome in HFpEF patients since it was demonstrated that biomarkers associated with the high-risk cluster were related to the immune system, signal transduction cascades, cell interactions and metabolismBiomarkers (and pathways) identified in this study may help select high-risk HFpEF patients which could be helpful for the inclusion/exclusion of patients in future trials.Our findings may be the basis of investigating therapies specifically targeting these pathways and the potential use of corresponding markers potentially identifying patients with distinct mechanistic bioprofiles most likely to respond to the selected mechanistically targeted therapies.

Machine Learning-Derived Echocardiographic Phenotypes Predict Heart Failure Incidence in Asymptomatic Individuals.

OBJECTIVES: This study sought to identify homogenous echocardiographic phenotypes in community-based cohorts and assess their association with outcomes. BACKGROUND: Asymptomatic cardiac dysfunction leads to a high risk of long-term cardiovascular morbidity and mortality; however, better echocardiographic classification of asymptomatic individuals remains a challenge. METHODS: Echocardiographic phenotypes were identified using K-means clustering in the first generation of the STANISLAS (Yearly non-invasive follow-up of Health status of Lorraine insured inhabitants) cohort (N = 827; mean age: 60 ± 5 years; men: 48%), and their associations with vascular function and circulating biomarkers were also assessed. These phenotypes were externally validated in the Malmö Preventive Project cohort (N = 1,394; mean age: 67 ± 6 years; men: 70%), and their associations with the composite of cardiovascular mortality (CVM) or heart failure hospitalization (HFH) were assessed as well. RESULTS: Three echocardiographic phenotypes were identified as "mostly normal (MN)" (n = 334), "diastolic changes (D)" (n = 323), and "diastolic changes with structural remodeling (D/S)" (n = 170). The D and D/S phenotypes had similar ages, body mass indices, cardiovascular risk factors, vascular impairments, and diastolic function changes. The D phenotype consisted mainly of women and featured increased levels of inflammatory biomarkers, whereas the D/S phenotype, consisted predominantly of men, displayed the highest values of left ventricular mass, volume, and remodeling biomarkers. The phenotypes were predicted based on a simple algorithm including e', left ventricular mass and volume (e'VM algorithm). In the Malmö cohort, subgroups derived from e'VM algorithm were significantly associated with a higher risk of CVM and HFH (adjusted HR in the D phenotype = 1.87; 95% CI: 1.04 to 3.37; adjusted HR in the D/S phenotype = 3.02; 95% CI: 1.71 to 5.34). CONCLUSIONS: Among asymptomatic, middle-aged individuals, echocardiographic data-driven classification based on the simple e'VM algorithm identified profiles with different long-term HF risk. (4th Visit at 17 Years of Cohort STANISLAS-Stanislas Ancillary Study ESCIF [STANISLASV4]; NCT01391442).

Head-to-head comparison of clustering methods for heterogeneous data: a simulation-driven benchmark.

The choice of the most appropriate unsupervised machine-learning method for "heterogeneous" or "mixed" data, i.e. with both continuous and categorical variables, can be challenging. Our aim was to examine the performance of various clustering strategies for mixed data using both simulated and real-life data. We conducted a benchmark analysis of "ready-to-use" tools in R comparing 4 model-based (Kamila algorithm, Latent Class Analysis, Latent Class Model [LCM] and Clustering by Mixture Modeling) and 5 distance/dissimilarity-based (Gower distance or Unsupervised Extra Trees dissimilarity followed by hierarchical clustering or Partitioning Around Medoids, K-prototypes) clustering methods. Clustering performances were assessed by Adjusted Rand Index (ARI) on 1000 generated virtual populations consisting of mixed variables using 7 scenarios with varying population sizes, number of clusters, number of continuous and categorical variables, proportions of relevant (non-noisy) variables and degree of variable relevance (low, mild, high). Clustering methods were then applied on the EPHESUS randomized clinical trial data (a heart failure trial evaluating the effect of eplerenone) allowing to illustrate the differences between different clustering techniques. The simulations revealed the dominance of K-prototypes, Kamila and LCM models over all other methods. Overall, methods using dissimilarity matrices in classical algorithms such as Partitioning Around Medoids and Hierarchical Clustering had a lower ARI compared to model-based methods in all scenarios. When applying clustering methods to a real-life clinical dataset, LCM showed promising results with regard to differences in (1) clinical profiles across clusters, (2) prognostic performance (highest C-index) and (3) identification of patient subgroups with substantial treatment benefit. The present findings suggest key differences in clustering performance between the tested algorithms (limited to tools readily available in R). In most of the tested scenarios, model-based methods (in particular the Kamila and LCM packages) and K-prototypes typically performed best in the setting of heterogeneous data.

Prognostic impact of plasma volume estimated from hemoglobin and hematocrit in heart failure with preserved ejection fraction.

BACKGROUND: Plasma volume (PV) estimated from Duarte's formula (based on hemoglobin/hematocrit) has been associated with poor prognosis in patients with heart failure (HF). There are, however, limited data regarding the association of estimated PV status (ePVS) derived from hemoglobin/hematocrit with clinical profiles and study outcomes in patients with HF and preserved ejection fraction (HFpEF). METHODS AND RESULTS: Patients from North and South America enrolled in the Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist trial (TOPCAT) with available hemoglobin/hematocrit data were studied. The association between ePVS (Duarte formula and Hakim formula) and the composite of cardiovascular mortality, HF hospitalization, or aborted cardiac arrest was assessed. Among 1747 patients (age 71.6 years; males 50.1%), mean ePVS derived from Duarte formula was 4.9 ± 1.0 mL/g. Higher Duarte-derived ePVS was associated with prior HF admission, diabetes, more severe congestion, poor renal function, higher natriuretic peptide level, and E/e'. After adjustment for potential covariates including natriuretic peptide, higher Duarte-derived ePVS was associated with an increased rate of the primary outcome [highest vs. lowest ePVS quartile: adjusted-HR (95%CI) = 1.79 (1.28-2.50), p < 0.001]. Duarte-derived ePVS improved prognostic performance on top of clinical and routine variables (including natriuretic peptides) (NRI = 11, p < 0.001), whereas Hakim-derived ePVS did not (p = 0.59). The prognostic value of Duarte-derived ePVS was not modified by renal function (P interaction > 0.10 for all outcomes). CONCLUSION: ePVS from Duarte's formula was associated with congestion status and improved risk stratification regardless of renal function. Our findings suggest that Duarte-derived ePVS is a useful congestion variable in patients with HFpEF.

Clinical determinants and prognostic implications of renin and aldosterone in patients with symptomatic heart failure.

AIMS: Activation of the renin-angiotensin-aldosterone system plays an important role in the pathophysiology of heart failure (HF) and has been associated with poor prognosis. There are limited data on the associations of renin and aldosterone levels with clinical profiles, treatment response, and study outcomes in patients with HF. METHODS AND RESULTS: We analysed 2,039 patients with available baseline renin and aldosterone levels in BIOSTAT-CHF (a systems BIOlogy study to Tailored Treatment in Chronic Heart Failure). The primary outcome was the composite of all-cause mortality or HF hospitalization. We also investigated changes in renin and aldosterone levels after administration of mineralocorticoid receptor antagonists (MRAs) in a subset of the EPHESUS trial and in an acute HF cohort (PORTO). In BIOSTAT-CHF study, median renin and aldosterone levels were 85.3 (percentile25-75 = 28-247) µIU/mL and 9.4 (percentile25-75 = 4.4-19.8) ng/dL, respectively. Prior HF admission, lower blood pressure, sodium, poorer renal function, and MRA treatment were associated with higher renin and aldosterone. Higher renin was associated with an increased rate of the primary outcome [highest vs. lowest renin tertile: adjusted-HR (95% CI) = 1.47 (1.16-1.86), P = 0.002], whereas higher aldosterone was not [highest vs. lowest aldosterone tertile: adjusted-HR (95% CI) = 1.16 (0.93-1.44), P = 0.19]. Renin and/or aldosterone did not improve the BIOSTAT-CHF prognostic models. The rise in aldosterone with the use of MRAs was observed in EPHESUS and PORTO studies. CONCLUSIONS: Circulating levels of renin and aldosterone were associated with both the disease severity and use of MRAs. By reflecting both the disease and its treatments, the prognostic discrimination of these biomarkers was poor. Our data suggest that the "point" measurement of renin and aldosterone in HF is of limited clinical utility.

Sex differences in circulating proteins in heart failure with preserved ejection fraction.

BACKGROUND: Many patients with heart failure with preserved ejection fraction (HFpEF) are women. Exploring mechanisms underlying the sex differences may improve our understanding of the pathophysiology of HFpEF. Studies focusing on sex differences in circulating proteins in HFpEF patients are scarce. METHODS: A total of 415 proteins were analyzed in 392 HFpEF patients included in The Metabolic Road to Diastolic Heart Failure: Diastolic Heart Failure study (MEDIA-DHF). Sex differences in these proteins were assessed using adjusted logistic regression analyses. The associations between candidate proteins and cardiovascular (CV) death or CV hospitalization (with sex interaction) were assessed using Cox regression models. RESULTS: We found 9 proteins to be differentially expressed between female and male patients. Women expressed more LPL and PLIN1, which are markers of lipid metabolism; more LHB, IGFBP3, and IL1RL2 as markers of transcriptional regulation; and more Ep-CAM as marker of hemostasis. Women expressed less MMP-3, which is a marker associated with extracellular matrix organization; less NRP1, which is associated with developmental processes; and less ACE2, which is related to metabolism. Sex was not associated with the study outcomes (adj. HR 1.48, 95% CI 0.83-2.63), p = 0.18. CONCLUSION: In chronic HFpEF, assessing sex differences in a wide range of circulating proteins led to the identification of 9 proteins that were differentially expressed between female and male patients. These findings may help further investigations into potential pathophysiological processes contributing to HFpEF.