Dataset Information

Comparing machine learning algorithms for multimorbidity prediction: An example from the Elsa-Brasil study.

ABSTRACT:

Background

Multimorbidity is a worldwide concern related to greater disability, worse quality of life, and mortality. The early prediction is crucial for preventive strategies design and integrative medical practice. However, knowledge about how to predict multimorbidity is limited, possibly due to the complexity involved in predicting multiple chronic diseases.

Methods

In this study, we present the use of a machine learning approach to build cost-effective multimorbidity prediction models. Based on predictors easily obtainable in clinical practice (sociodemographic, clinical, family disease history and lifestyle), we build and compared the performance of seven multilabel classifiers (multivariate random forest, and classifier chain, binary relevance and binary dependence, with random forest and support vector machine as base classifiers), using a sample of 15105 participants from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). We developed a web application for the building and use of prediction models.

Results

Classifier chain with random forest as base classifier performed better (accuracy = 0.34, subset accuracy = 0.15, and Hamming Loss = 0.16). For different feature sets, random forest based classifiers outperformed those based on support vector machine. BMI, blood pressure, sex, and age were the features most relevant to multimorbidity prediction.

Conclusions

Our results support the choice of random forest based classifiers for multimorbidity prediction.

SUBMITTER: Polessa Paula D

PROVIDER: S-EPMC9543987 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Comparing machine learning algorithms for multimorbidity prediction: An example from the Elsa-Brasil study.

Polessa Paula Daniela D Barbosa Aguiar Odaleia O Pruner Marques Larissa L Bensenor Isabela I Suemoto Claudia Kimie CK Mendes da Fonseca Maria de Jesus MJ Griep Rosane Härter RH

PloS one 20221007 10

<h4>Background</h4>Multimorbidity is a worldwide concern related to greater disability, worse quality of life, and mortality. The early prediction is crucial for preventive strategies design and integrative medical practice. However, knowledge about how to predict multimorbidity is limited, possibly due to the complexity involved in predicting multiple chronic diseases.<h4>Methods</h4>In this study, we present the use of a machine learning approach to build cost-effective multimorbidity predicti ...[more]

PMID: 36206287

Similar Datasets

Project description:BackgroundEvidence of multimorbidity has come mainly from high-income regions, while disparities among racial groups have been less explored. This study examined racial differences in multimorbidity in the multiracial cohort of the Longitudinal Study of Adult Health (Estudo Longitudinal de Saúde do Adulto), ELSA-Brasil.MethodsThe study examined baseline (2008-2010) data for 14 099 ELSA-Brasil participants who self-reported being white, mixed-race, or black. A list of 16 morbidities was used to evaluate multimorbidity, operationalised by simple count into ≥ 2, ≥ 3, ≥ 4, ≥ 5 and ≥ 6 morbidities, in addition to evaluating the number of coexisting conditions. Prevalence ratios (PR) were estimated from logistic models and a quantile model was used to examine racial differences graphically in the distribution quantiles for the number of morbidities.ResultsOverall prevalence of multimorbidity (≥ 2 morbidities) was 70% and, after controlling for age and sex, was greater among mixed-race and black participants - by 6% (PR: 1.06; 95% CI: 1.03-1.08) and 9% (PR: 1.09; 95% CI: 1.06-1.12), respectively - than among white participants. As the cutoff value for defining multimorbidity was raised, so the strength of the association increased, especially among blacks: if set at ≥ 6 morbidities, the prevalence was 27% greater for those of mixed-race (PR: 1.27; 95% CI: 1.07-1.50) and 47% greater for blacks (PR: 1.47; 95% CI: 1.22-1.76) than for whites. The disparities were smaller in the lower morbidity distribution quantiles and larger in the upper quantiles, indicating a heavier burden of disease, particularly on blacks.ConclusionsMultimorbidity was common among adults and older adults in a Brazilian cohort, but important racial inequalities were found. Raising the cutoff point for defining multimorbidity revealed stronger associations between race/skin colour and multimorbidity, indicating a higher prevalence of multimorbidity among mixed-race and black individuals than among whites and that the former groups coexisted more often with more complex health situations (with more coexisting morbidities). Interventions to prevent and manage the condition of multimorbidity that consider the social determinants of health and historically discriminated populations in low- and middle-income regions are necessary.

Project description:IntroductionOsteoporosis, characterized by low bone mineral density (BMD), is an increasingly serious public health issue. So far, several traditional regression models and machine learning (ML) algorithms have been proposed for predicting osteoporosis risk. However, these models have shown relatively low accuracy in clinical implementation. Recently proposed deep learning (DL) approaches, such as deep neural network (DNN), which can discover knowledge from complex hidden interactions, offer a new opportunity to improve predictive performance. In this study, we aimed to assess whether DNN can achieve a better performance in osteoporosis risk prediction.MethodsBy utilizing hip BMD and extensive demographic and routine clinical data of 8,134 subjects with age more than 40 from the Louisiana Osteoporosis Study (LOS), we developed and constructed a novel DNN framework for predicting osteoporosis risk and compared its performance in osteoporosis risk prediction with four conventional ML models, namely random forest (RF), artificial neural network (ANN), k-nearest neighbor (KNN), and support vector machine (SVM), as well as a traditional regression model termed osteoporosis self-assessment tool (OST). Model performance was assessed by area under 'receiver operating curve' (AUC) and accuracy.ResultsBy using 16 discriminative variables, we observed that the DNN approach achieved the best predictive performance (AUC = 0.848) in classifying osteoporosis (hip BMD T-score ≤ -1.0) and non-osteoporosis risk (hip BMD T-score > -1.0) subjects, compared to the other approaches. Feature importance analysis showed that the top 10 most important variables identified by the DNN model were weight, age, gender, grip strength, height, beer drinking, diastolic pressure, alcohol drinking, smoke years, and economic level. Furthermore, we performed subsampling analysis to assess the effects of varying number of sample size and variables on the predictive performance of these tested models. Notably, we observed that the DNN model performed equally well (AUC = 0.846) even by utilizing only the top 10 most important variables for osteoporosis risk prediction. Meanwhile, the DNN model can still achieve a high predictive performance (AUC = 0.826) when sample size was reduced to 50% of the original dataset.ConclusionIn conclusion, we developed a novel DNN model which was considered to be an effective algorithm for early diagnosis and intervention of osteoporosis in the aging population.

Project description:Many chemicals that disrupt endocrine function have been linked to a variety of adverse biological outcomes. However, screening for endocrine disruption using in vitro or in vivo approaches is costly and time-consuming. Computational methods, e.g., quantitative structure-activity relationship models, have become more reliable due to bigger training sets, increased computing power, and advanced machine learning algorithms, such as multilayered artificial neural networks. Machine learning models can be used to predict compounds for endocrine disrupting capabilities, such as binding to the estrogen receptor (ER), and allow for prioritization and further testing. In this work, an exhaustive comparison of multiple machine learning algorithms, chemical spaces, and evaluation metrics for ER binding was performed on public data sets curated using in-house cheminformatics software (Assay Central). Chemical features utilized in modeling consisted of binary fingerprints (ECFP6, FCFP6, ToxPrint, or MACCS keys) and continuous molecular descriptors from RDKit. Each feature set was subjected to classic machine learning algorithms (Bernoulli Naive Bayes, AdaBoost Decision Tree, Random Forest, Support Vector Machine) and Deep Neural Networks (DNN). Models were evaluated using a variety of metrics: recall, precision, F1-score, accuracy, area under the receiver operating characteristic curve, Cohen's Kappa, and Matthews correlation coefficient. For predicting compounds within the training set, DNN has an accuracy higher than that of other methods; however, in 5-fold cross validation and external test set predictions, DNN and most classic machine learning models perform similarly regardless of the data set or molecular descriptors used. We have also used the rank normalized scores as a performance-criteria for each machine learning method, and Random Forest performed best on the validation set when ranked by metric or by data sets. These results suggest classic machine learning algorithms may be sufficient to develop high quality predictive models of ER activity.

Project description:BackgroundThe temporal relationships across cardiometabolic diseases (CMDs) were recently conceptualized as the cardiometabolic continuum (CMC), sequence of cardiovascular events that stem from gene-environmental interactions, unhealthy lifestyle influences, and metabolic diseases such as diabetes, and hypertension. While the physiological pathways linking metabolic and cardiovascular diseases have been investigated, the study of the sex and population differences in the CMC have still not been described.MethodsWe present a machine learning approach to model the CMC and investigate sex and population differences in two distinct cohorts: the UK Biobank (17,700 participants) and the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) (7162 participants). We consider the following CMDs: hypertension (Hyp), diabetes (DM), heart diseases (HD: angina, myocardial infarction, or heart failure), and stroke (STK). For the identification of the CMC patterns, individual trajectories with the time of disease occurrence were clustered using k-means. Based on clinical, sociodemographic, and lifestyle characteristics, we built multiclass random forest classifiers and used the SHAP methodology to evaluate feature importance.ResultsFive CMC patterns were identified across both sexes and cohorts: EarlyHyp, FirstDM, FirstHD, Healthy, and LateHyp, named according to prevalence and disease occurrence time that depicted around 95%, 78%, 75%, 88% and 99% of individuals, respectively. Within the UK Biobank, more women were classified in the Healthy cluster and more men in all others. In the EarlyHyp and LateHyp clusters, isolated hypertension occurred earlier among women. Smoking habits and education had high importance and clear directionality for both sexes. For ELSA-Brasil, more men were classified in the Healthy cluster and more women in the FirstDM. The diabetes occurrence time when followed by hypertension was lower among women. Education and ethnicity had high importance and clear directionality for women, while for men these features were smoking, alcohol, and coffee consumption.ConclusionsThere are clear sex differences in the CMC that varied across the UK and Brazilian cohorts. In particular, disadvantages regarding incidence and the time to onset of diseases were more pronounced in Brazil, against woman. The results show the need to strengthen public health policies to prevent and control the time course of CMD, with an emphasis on women.

Project description:BackgroundTrauma is one of the most critical public health issues worldwide, leading to death and disability and influencing all age groups. Therefore, there is great interest in models for predicting mortality in trauma patients admitted to the ICU. The main objective of the present study is to develop and evaluate SMOTE-based machine-learning tools for predicting hospital mortality in trauma patients with imbalanced data.MethodsThis retrospective cohort study was conducted on 126 trauma patients admitted to an intensive care unit at Besat hospital in Hamadan Province, western Iran, from March 2020 to March 2021. Data were extracted from the medical information records of patients. According to the imbalanced property of the data, SMOTE techniques, namely SMOTE, Borderline-SMOTE1, Borderline-SMOTE2, SMOTE-NC, and SVM-SMOTE, were used for primary preprocessing. Then, the Decision Tree (DT), Random Forest (RF), Naive Bayes (NB), Artificial Neural Network (ANN), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost) methods were used to predict patients' hospital mortality with traumatic injuries. The performance of the methods used was evaluated by sensitivity, specificity, Positive Predictive Value (PPV), Negative Predictive Value (NPV), accuracy, Area Under the Curve (AUC), Geometric Mean (G-means), F1 score, and P-value of McNemar's test.ResultsOf the 126 patients admitted to an ICU, 117 (92.9%) survived and 9 (7.1%) died. The mean follow-up time from the date of trauma to the date of outcome was 3.98 ± 4.65 days. The performance of ML algorithms is not good with imbalanced data, whereas the performance of SMOTE-based ML algorithms is significantly improved. The mean area under the ROC curve (AUC) of all SMOTE-based models was more than 91%. F1-score and G-means before balancing the dataset were below 70% for all ML models except ANN. In contrast, F1-score and G-means for the balanced datasets reached more than 90% for all SMOTE-based models. Among all SMOTE-based ML methods, RF and ANN based on SMOTE and XGBoost based on SMOTE-NC achieved the highest value for all evaluation criteria.ConclusionsThis study has shown that SMOTE-based ML algorithms better predict outcomes in traumatic injuries than ML algorithms. They have the potential to assist ICU physicians in making clinical decisions.

Project description:BackgroundPrevention and reduction of excessive use of alcohol represents damages to society in general. In turn, arterial hypertension is the main attributable risk factor premature life lost years and disability.ObjectiveTo investigate the relationship between alcohol consumption and high blood pressure in participants of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).MethodologyA baseline data of total of 7,655 participants volunteers between 35 and 74 years of age, of both genders, in six educational and research institutions of three different regions of the country were interviewed between 2008-2010. Socioeconomic, haemodynamic, anthropometric and health data were collected in the research centers of ELSA-Brasil. The presence of high blood pressure was identified when the systolic blood pressure was ≥140 mm Hg and/or the diastolic was ≥90 mm Hg. Alcohol consumption was estimated and categorized regarding consumption and pattern of ingestion. The Student's t-test, chi-squared and logistic regression tests were used for analysis, including potential co-variables of the model, and a 5% significance level was adopted.ResultsA dose-response relation was observed for the consumption of alcohol (g/week) in systolic blood pressure and diastolic blood pressure. Alcohol consumption was associated with high blood pressure in men who reported moderate (OR = 1.69; 95%CI 1.35-2.11) and excessive (OR = 2.70; 95%CI 2.04-3.59) consumption. Women have nearly three times more chance of presenting elevated blood pressure when presenting excessive consumption (OR = 2.86, 95%CI 1.77-4.63), and binge drinkers who drink more than 2 to 3 times a month have approximately 70% more chance of presenting with elevated blood pressure, after adjusting for consumption of drinks with meals.ConclusionThe consumption of alcohol beverages increases the odds of elevated blood pressure, especially among excessive drinkers. Therefore alcohol consumption needs a more robust regulation in view of its impact on population health.

Dataset Information

Comparing machine learning algorithms for multimorbidity prediction: An example from the Elsa-Brasil study.

Background

Methods

Results

Conclusions

Publications

Comparing machine learning algorithms for multimorbidity prediction: An example from the Elsa-Brasil study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets