Using machine learning to predict COVID-19 infection and severity risk among 4,510 aged adults: a UK Biobank cohort study.
Ontology highlight
ABSTRACT: Background
Many risk factors have emerged for novel 2019 coronavirus disease (COVID-19). It is relatively unknown how these factors collectively predict COVID-19 infection risk, as well as risk for a severe infection (i.e., hospitalization).Methods
Among aged adults (69.3 ± 8.6 years) in UK Biobank, COVID-19 data was downloaded for 4,510 participants with 7,539 test cases. We downloaded baseline data from 10-14 years ago, including demographics, biochemistry, body mass, and other factors, as well as antibody titers for 20 common to rare infectious diseases. Permutation-based linear discriminant analysis was used to predict COVID-19 risk and hospitalization risk. Probability and threshold metrics included receiver operating characteristic curves to derive area under the curve (AUC), specificity, sensitivity, and quadratic mean.Results
The "best-fit" model for predicting COVID-19 risk achieved excellent discrimination (AUC=0.969, 95% CI=0.934-1.000). Factors included age, immune markers, lipids, and serology titers to common pathogens like human cytomegalovirus. The hospitalization "best-fit" model was more modest (AUC=0.803, 95% CI=0.663-0.943) and included only serology titers.Conclusions
Accurate risk profiles can be created using standard self-report and biomedical data collected in public health and medical settings. It is also worthwhile to further investigate if prior host immunity predicts current host immunity to COVID-19.
SUBMITTER: Willette AA
PROVIDER: S-EPMC7302228 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Many risk factors have emerged for novel 2019 coronavirus disease (COVID-19). It is relatively unknown how these factors collectively predict COVID-19 infection risk, as well as risk for a severe infection (i.e., hospitalization). Among aged adults (69.3 ± 8.6 years) in UK Biobank, COVID-19 data was downloaded for 4510 participants with 7539 test cases. We downloaded baseline data from 10 to 14 years ago, including demographics, biochemistry, body mass, and other factors, as well as antibody titers for 20 common to rare infectious diseases in a subset of 80 participants with 124 test cases. Permutation-based linear discriminant analysis was used to predict COVID-19 risk and hospitalization risk. Probability and threshold metrics included receiver operating characteristic curves to derive area under the curve (AUC), specificity, sensitivity, and quadratic mean. Model predictions using the full cohort were marginal. The "best-fit" model for predicting COVID-19 risk was found in the subset of participants with antibody titers, which achieved excellent discrimination (AUC 0.969, 95% CI 0.934-1.000). Factors included age, immune markers, lipids, and serology titers to common pathogens like human cytomegalovirus. The hospitalization "best-fit" model was more modest (AUC 0.803, 95% CI 0.663-0.943) and included only serology titers, again in the subset group. Accurate risk profiles can be created using standard self-report and biomedical data collected in public health and medical settings. It is also worthwhile to further investigate if prior host immunity predicts current host immunity to COVID-19.
| S-EPMC9092926 | biostudies-literature
Project description:BackgroundSuicidal behaviors, including suicide deaths and attempts, are major public health concerns. However, previous suicide models required a huge amount of input features, resulting in limited applicability in clinical practice.ObjectiveWe aimed to construct applicable models (ie, with limited features) for short- and long-term suicidal behavior prediction. We further validated these models among individuals with different genetic risks of suicide.MethodsBased on the prospective cohort of UK Biobank, we included 223 (0.06%) eligible cases of suicide attempts or deaths, according to hospital inpatient or death register data within 1 year from baseline and randomly selected 4460 (1.18%) controls (1:20) without such records. We similarly identified 833 (0.22%) cases of suicidal behaviors 1 to 6 years from baseline and 16,660 (4.42%) corresponding controls. Based on 143 input features, mainly including sociodemographic, environmental, and psychosocial factors; medical history; and polygenic risk scores (PRS) for suicidality, we applied a bagged balanced light gradient-boosting machine (LightGBM) with stratified 10-fold cross-validation and grid-search to construct the full prediction models for suicide attempts or deaths within 1 year or between 1 and 6 years. The Shapley Additive Explanations (SHAP) approach was used to quantify the importance of input features, and the top 20 features with the highest SHAP values were selected to train the applicable models. The external validity of the established models was assessed among 50,310 individuals who participated in UK Biobank repeated assessments both overall and by the level of PRS for suicidality.ResultsIndividuals with suicidal behaviors were on average 56 years old, with equal sex distribution. The application of these full models in the external validation data set demonstrated good model performance, with the area under the receiver operating characteristic (AUROC) curves of 0.919 and 0.892 within 1 year and between 1 and 6 years, respectively. Importantly, the applicable models with the top 20 most important features showed comparable external-validated performance (AUROC curves of 0.901 and 0.885) as the full models, based on which we found that individuals in the top quintile of predicted risk accounted for 91.7% (n=11) and 80.7% (n=25) of all suicidality cases within 1 year and during 1 to 6 years, respectively. We further obtained comparable prediction accuracy when applying these models to subpopulations with different genetic susceptibilities to suicidality. For example, for the 1-year risk prediction, the AUROC curves were 0.907 and 0.885 for the high (>2nd tertile of PRS) and low (<1st) genetic susceptibilities groups, respectively.ConclusionsWe established applicable machine learning-based models for predicting both the short- and long-term risk of suicidality with high accuracy across populations of varying genetic risk for suicide, highlighting a cost-effective method of identifying individuals with a high risk of suicidality.
| S-EPMC9989910 | biostudies-literature
Project description:BackgroundMachine learning (ML) models can be used to predict future frailty in the community setting. However, outcome variables for epidemiologic data sets such as frailty usually have an imbalance between categories, that is, there are far fewer individuals classified as frail than as nonfrail, adversely affecting the performance of ML models when predicting the syndrome.MethodsA retrospective cohort study with participants (50 years or older) from the English Longitudinal Study of Ageing who were nonfrail at baseline (2008-2009) and reassessed for the frailty phenotype at 4-year follow-up (2012-2013). Social, clinical, and psychosocial baseline predictors were selected to predict frailty at follow-up in ML models (Logistic Regression, Random Forest [RF], Support Vector Machine, Neural Network, K-nearest neighbor, and Naive Bayes classifier).ResultsOf all the 4 378 nonfrail participants at baseline, 347 became frail at follow-up. The proposed combined oversampling and undersampling method to adjust imbalanced data improved the performance of the models, and RF had the best performance, with areas under the receiver-operating characteristic curve and the precision-recall curve of 0.92 and 0.97, respectively, specificity of 0.83, sensitivity of 0.88, and balanced accuracy of 85.5% for balanced data. Age, chair-rise test, household wealth, balance problems, and self-rated health were the most important frailty predictors in most of the models trained with balanced data.ConclusionsML proved useful in identifying individuals who became frail over time, and this result was made possible by balancing the data set. This study highlighted factors that may be useful in the early detection of frailty.
| S-EPMC10613015 | biostudies-literature
Project description:Background Nonalcoholic fatty liver disease (NAFLD) is the most prevalent liver disease worldwide. Cardiovascular disease (CVD) is the leading cause of mortality among patients with NAFLD. The aim of our study was to develop a machine learning algorithm integrating clinical, lifestyle, and genetic risk factors to identify CVD in patients with NAFLD. Methods and Results We created a cohort of patients with NAFLD from the UK Biobank, diagnosed according to proton density fat fraction from magnetic resonance imaging data sets. A total of 400 patients with NAFLD with subclinical atherosclerosis or clinical CVD, defined by disease codes, constituted cases and 446 NAFLD cases with no CVD constituted controls. We evaluated 7 different supervised machine learning approaches on clinical, lifestyle, and genetic variables for identifying CVD in patients with NAFLD. The most significant clinical and lifestyle variables observed by the predictive modeling were age (59 years [54.00-63.00 years]), hypertension (145 mm Hg [134.0-156.0 mm Hg] and 85 mm Hg [79.00-93.00 mm Hg]), waist circumference (98 cm [95.00-105.00 cm]), and sedentary lifestyle, defined as time spent watching TV >4 h/d. In the genetic data, single-nucleotide polymorphisms in IL16 and ANKLE1 gene were most significant. Our proposed ensemble-based integrative machine learning model achieved an area under the curve of 0.849 using the random forest modeling for CVD prediction. Conclusions We propose a machine learning algorithm that identifies CVD in patients with NAFLD through integration of significant clinical, lifestyle, and genetic risk factors. These patients with NAFLD at higher risk of CVD should be flagged for screening and aggressive treatment of their cardiometabolic risk factors to prevent cardiovascular morbidity and mortality.
| S-EPMC9075189 | biostudies-literature
Project description:ObjectiveAtrial fibrillation (AF) is the most common cardiac arrythmia, and it is associated with increased risk for ischemic stroke, which is underestimated, as AF can be asymptomatic. The aim of this study was to develop optimal ML models for prediction of AF in the population, and secondly for ischemic stroke in AF patients.MethodsTo develop ML models for prediction of 1) AF in the general population and 2) ischemic stroke in patients with AF we constructed XGBoost, LightGBM, Random Forest, Deep Neural Network, Support Vector Machine and Lasso penalised logistic regression models using UK-Biobank's extensive real-world clinical data, questionnaires, as well as biochemical and genetic data, and their predictive performances were compared. Ranking and contribution of the different features was assessed by SHapley Additive exPlanations (SHAP) analysis. The clinical tool CHA2DS2-VASc for prediction of ischemic stroke among AF patients, was used for comparison to the best performing ML model.FindingsThe best performing model for AF prediction was LightGBM, with an area-under-the-roc-curve (AUROC) of 0.729 (95% confidence intervals (CI): 0.719, 0.738). The best performing model for ischemic stroke prediction in AF patients was XGBoost with AUROC of 0.631 (95% CI: 0.604, 0.657). The improved AUROC in the XGBoost model compared to CHA2DS2-VASc was statistically significant based on DeLong's test (p-value = 2.20E-06). In addition, the SHAP analysis showed that several peripheral blood biomarkers (e.g. creatinine, glycated haemoglobin, monocytes) were associated with ischemic stroke, which are not considered by CHA2DS2-VASc.ImplicationsThe best performing ML models presented have the potential for clinical use, but further validation in independent studies is required. Our results endorse the incorporation of some routinely measured blood biomarkers for ischemic stroke prediction in AF patients.
| S-EPMC10987914 | biostudies-literature
Project description:BackgroundIt is now well recognised that the risk of severe COVID-19 increases with some long-term conditions (LTCs). However, prior research primarily focuses on individual LTCs and there is a lack of data on the influence of multimorbidity (≥2 LTCs) on the risk of COVID-19. Given the high prevalence of multimorbidity, more detailed understanding of the associations with multimorbidity and COVID-19 would improve risk stratification and help protect those most vulnerable to severe COVID-19. Here we examine the relationships between multimorbidity, polypharmacy (a proxy of multimorbidity), and COVID-19; and how these differ by sociodemographic, lifestyle, and physiological prognostic factors.Methods and findingsWe studied data from UK Biobank (428,199 participants; aged 37-73; recruited 2006-2010) on self-reported LTCs, medications, sociodemographic, lifestyle, and physiological measures which were linked to COVID-19 test data. Poisson regression models examined risk of COVID-19 by multimorbidity/polypharmacy and effect modification by COVID-19 prognostic factors (age/sex/ethnicity/socioeconomic status/smoking/physical activity/BMI/systolic blood pressure/renal function). 4,498 (1.05%) participants were tested; 1,324 (0.31%) tested positive for COVID-19. Compared with no LTCs, relative risk (RR) of COVID-19 in those with 1 LTC was no higher (RR 1.12 (CI 0.96-1.30)), whereas those with ≥2 LTCs had 48% higher risk; RR 1.48 (1.28-1.71). Compared with no cardiometabolic LTCs, having 1 and ≥2 cardiometabolic LTCs had a higher risk of COVID-19; RR 1.28 (1.12-1.46) and 1.77 (1.46-2.15), respectively. Polypharmacy was associated with a dose response higher risk of COVID-19. All prognostic factors were associated with a higher risk of COVID-19 infection in multimorbidity; being non-white, most socioeconomically deprived, BMI ≥40 kg/m2, and reduced renal function were associated with the highest risk of COVID-19 infection: RR 2.81 (2.09-3.78); 2.79 (2.00-3.90); 2.66 (1.88-3.76); 2.13 (1.46-3.12), respectively. No multiplicative interaction between multimorbidity and prognostic factors was identified. Important limitations include the low proportion of UK Biobank participants with COVID-19 test data (1.05%) and UK Biobank participants being more affluent, healthier and less ethnically diverse than the general population.ConclusionsIncreasing multimorbidity, especially cardiometabolic multimorbidity, and polypharmacy are associated with a higher risk of developing COVID-19. Those with multimorbidity and additional factors, such as non-white ethnicity, are at heightened risk of COVID-19.
| S-EPMC7440632 | biostudies-literature
Project description:Mood disorders, including depression and anxiety, affect almost one-fifth of the world's adult population and are becoming increasingly prevalent. Mutations in circadian clock genes have previously been associated with mood disorders both directly and indirectly through alterations in circadian phase, suggesting that the circadian clock influences multiple molecular pathways involved in mood. By targeting previously identified single nucleotide polymorphisms (SNPs) that have been implicated in anxiety and depressive disorders, we use a combination of statistical and machine learning techniques to investigate associations with the generalized anxiety disorder assessment (GAD-7) scores in a UK Biobank sample of 90,882 individuals. As in previous studies, we observed that females exhibited higher GAD-7 scores than males regardless of genotype. Interestingly, we found no significant effects on anxiety from individual circadian gene variants; only circadian genotypes with multiple SNP variants showed significant associations with anxiety. For both sexes, severe anxiety is associated with a 120-fold increase in odds for individuals with CRY2_AG(rs1083852)/ZBTB20_TT(rs1394593) genotypes and is associated with a near 40-fold reduction in odds for individuals with PER3-A_CG(rs228697)/ZBTB20_TT(rs1394593) genotypes. We also report several sex-specific associations with anxiety. In females, the CRY2/ZBTB20 genotype combination showed a > 200-fold increase in odds of anxiety and PER3/ZBTB20 and CRY1 /PER3-A genotype combinations also appeared as female risk factors. In males, CRY1/PER3-A and PER3-B/ZBTB20 genotype combinations were associated with anxiety risk. Mediation analysis revealed direct associations of CRY2/ZBTB20 variant genotypes with moderate anxiety in females and CRY1/PER3-A variant genotypes with severe anxiety in males. The association of CRY1/PER3-A variant genotypes with severe anxiety in females was partially mediated by extreme evening chronotype. Our results reinforce existing findings that females exhibit stronger anxiety outcomes than males, and provide evidence for circadian gene associations with anxiety, particularly in females. Our analyses only identified significant associations using two-gene combinations, underscoring the importance of combined gene effects on anxiety risk. We describe novel, robust associations between gene combinations involving the ZBTB20 SNP (rs1394593) and risk of anxiety symptoms in a large population sample. Our findings also support previous findings that the ZBTB20 SNP is an important factor in mood disorders, including seasonal affective disorder. Our results suggest that reduced expression of this gene significantly modulates the risk of anxiety symptoms through direct influences on mood-related pathways. Together, these observations provide novel links between the circadian clockwork and anxiety symptoms and identify potential molecular pathways through which clock genes may influence anxiety risk.
| S-EPMC10724169 | biostudies-literature
Project description:Disinhibition is a prominent feature of multiple psychiatric disorders, and has been associated with poor long-term somatic outcomes. Modifiable lifestyle factors including diet and moderate-to-vigorous physical activity (MVPA) may be associated with disinhibition, but their contributions have not previously been quantified among middle-aged/older adults. Here, among N = 157,354 UK Biobank participants aged 40-69, we extracted a single disinhibition principal component and four dietary components (prudent diet, elimination of wheat/dairy/eggs, meat consumption, full-cream dairy consumption). In addition, latent profile analysis assigned participants to one of five empirical dietary groups: prudent-moderate, unhealthy, restricted, meat-avoiding, low-fat dairy. Disinhibition was regressed on the four dietary components, the dietary grouping variable, and self-reported MVPA. In men and women, disinhibition was negatively associated with prudent diet, and positively associated with wheat/dairy/eggs elimination. In men, disinhibition was also associated with consumption of meat and full-cream dairy products. Comparing groups, disinhibition was lower in the prudent-moderate diet (reference) group compared to all other groups. Absolute βs ranged from 0.02-0.13, indicating very weak effects. Disinhibition was not associated with MVPA. In conclusion, disinhibition is associated with multiple features of diet among middle-aged/older adults. Our findings foster specific hypotheses (e.g., early malnutrition, elevated immune-response) to be tested in alternative study designs.
| S-EPMC8151887 | biostudies-literature
Project description:Machine learning (ML) holds promise for precision psychiatry, but its predictive performance is unclear. We assessed whether ML provided added value over logistic regression for prediction of schizophrenia, and compared models built using polygenic risk scores (PRS) or clinical/demographic factors. LASSO and ridge-penalised logistic regression, support vector machines (SVM), random forests, boosting, neural networks and stacked models were trained to predict schizophrenia, using PRS for schizophrenia (PRSSZ), sex, parental depression, educational attainment, winter birth, handedness and number of siblings as predictors. Models were evaluated for discrimination using area under the receiver operator characteristic curve (AUROC) and relative importance of predictors using permutation feature importance (PFI). In a secondary analysis, fitted models were tested for association with schizophrenia-related traits which had not been used in model development. Following learning curve analysis, 738 cases and 3690 randomly sampled controls were selected from the UK Biobank. ML models combining all predictors showed the highest discrimination (linear SVM, AUROC = 0.71), but did not significantly outperform logistic regression. AUROC was robust over 100 random resamples of controls. PFI identified PRSSZ as the most important predictor. Highest variance in fitted models was explained by schizophrenia-related traits including fluid intelligence (most associated: linear SVM), digit symbol substitution (RBF SVM), BMI (XGBoost), smoking status (XGBoost) and deprivation (linear SVM). In conclusion, ML approaches did not provide substantial added value for prediction of schizophrenia over logistic regression, as indexed by AUROC; however, risk scores derived with different ML approaches differ with respect to association with schizophrenia-related traits.
| S-EPMC9399753 | biostudies-literature
Project description:The effect of gene expression data on diagnosis remains limited. Here, we show how diagnosis and classification of growth hormone deficiency (GHD) can be achieved from a single blood sample using a combination of transcriptomics and random forest analysis. Prepubertal treatment-naive children with GHD (n = 98) were enrolled from the PREDICT study, and controls (n = 26) were acquired from online data sets. Whole blood gene expression was correlated with peak growth hormone (GH) using rank regression and a random forest algorithm tested for prediction of the presence of GHD and in classification of GHD as severe (peak GH <4 μg/l) and nonsevere (peak ≥4 μg/l). Performance was assessed using area under the receiver operating characteristic curve (AUC-ROC). Rank regression identified 347 probe sets in which gene expression correlated with peak GH concentrations (r = ± 0.28, P < 0.01). These 347 probe sets yielded an AUC-ROC of 0.95 for prediction of GHD status versus controls and an AUC-ROC of 0.93 for prediction of GHD severity. This study demonstrates highly accurate diagnosis and disease classification for GHD using a combination of transcriptomics and random forest analysis. NCT00256126 and NCT00699855. Merck and the National Institute for Health Research (CL-2012-06-005).
| S-EPMC5928867 | biostudies-literature