Comparing the accuracy of several network-based COVID-19 prediction algorithms.
Ontology highlight
ABSTRACT:
SUBMITTER: Achterberg MA
PROVIDER: S-EPMC7546239 | biostudies-literature | 2020 Oct
REPOSITORIES: biostudies-literature
Publications
Comparing the accuracy of several network-based COVID-19 prediction algorithms.
Achterberg Massimo A MA Prasse Bastian B Ma Long L Trajanovski Stojan S Kitsak Maksim M Van Mieghem Piet P
International journal of forecasting 20201009 2
Researchers from various scientific disciplines have attempted to forecast the spread of coronavirus disease 2019 (COVID-19). The proposed epidemic prediction methods range from basic curve fitting methods and traffic interaction models to machine-learning approaches. If we combine all these approaches, we obtain the Network Inference-based Prediction Algorithm (NIPA). In this paper, we analyse a diverse set of COVID-19 forecast algorithms, including several modifications of NIPA. Among the algo ...[more]
Similar Datasets
Project description:As predicting the trajectory of COVID-19 is challenging, machine learning models could assist physicians in identifying high-risk individuals. This study compares the performance of 18 machine learning algorithms for predicting ICU admission and mortality among COVID-19 patients. Using COVID-19 patient data from the Mass General Brigham (MGB) Healthcare database, we developed and internally validated models using patients presenting to the Emergency Department (ED) between March-April 2020 (n = 3597) and further validated them using temporally distinct individuals who presented to the ED between May-August 2020 (n = 1711). We show that ensemble-based models perform better than other model types at predicting both 5-day ICU admission and 28-day mortality from COVID-19. CRP, LDH, and O2 saturation were important for ICU admission models whereas eGFR <60 ml/min/1.73 m2, and neutrophil and lymphocyte percentages were the most important variables for predicting mortality. Implementing such models could help in clinical decision-making for future infectious disease outbreaks including COVID-19.
| S-EPMC8140139 | biostudies-literature
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.
| S-EPMC10122327 | biostudies-literature
Project description:BackgroundThe severity of coronavirus (COVID-19) in patients with chronic comorbidities is much higher than in other patients, which can lead to their death. Machine learning (ML) algorithms as a potential solution for rapid and early clinical evaluation of the severity of the disease can help in allocating and prioritizing resources to reduce mortality.ObjectiveThe objective of this study was to predict the mortality risk and length of stay (LoS) of patients with COVID-19 and history of chronic comorbidities using ML algorithms.MethodsThis retrospective study was conducted by reviewing the medical records of COVID-19 patients with a history of chronic comorbidities from March 2020 to January 2021 in Afzalipour Hospital in Kerman, Iran. The outcome of patients, hospitalization was recorded as discharge or death. The filtering technique used to score the features and well-known ML algorithms were applied to predict the risk of mortality and LoS of patients. Ensemble Learning methods is also used. To evaluate the performance of the models, different measures including F1, precision, recall, and accuracy were calculated. The TRIPOD guideline assessed transparent reporting.ResultsThis study was performed on 1291 patients, including 900 alive and 391 dead patients. Shortness of breath (53.6%), fever (30.1%), and cough (25.3%) were the three most common symptoms in patients. Diabetes mellitus(DM) (31.3%), hypertension (HTN) (27.3%), and ischemic heart disease (IHD) (14.2%) were the three most common chronic comorbidities of patients. Twenty-six important factors were extracted from each patient's record. Gradient boosting model with 84.15% accuracy was the best model for predicting mortality risk and multilayer perceptron (MLP) with rectified linear unit function (MSE = 38.96) was the best model for predicting the LoS. The most common chronic comorbidities among these patients were DM (31.3%), HTN (27.3%), and IHD (14.2%). The most important factors in predicting the risk of mortality were hyperlipidemia, diabetes, asthma, and cancer, and in predicting LoS was shortness of breath.ConclusionThe results of this study showed that the use of ML algorithms can be a good tool to predict the risk of mortality and LoS of patients with COVID-19 and chronic comorbidities based on physiological conditions, symptoms, and demographic information of patients. The Gradient boosting and MLP algorithms can quickly identify patients at risk of death or long-term hospitalization and notify physicians to do appropriate interventions.
| S-EPMC10259141 | biostudies-literature
Project description:At the moment of writing, the future evolution of the COVID-19 epidemic is unclear. Predictions of the further course of the epidemic are decisive to deploy targeted disease control measures. We consider a network-based model to describe the COVID-19 epidemic in the Hubei province. The network is composed of the cities in Hubei and their interactions (e.g., traffic flow). However, the precise interactions between cities is unknown and must be inferred from observing the epidemic. We propose the Network-Inference-Based Prediction Algorithm (NIPA) to forecast the future prevalence of the COVID-19 epidemic in every city. Our results indicate that NIPA is beneficial for an accurate forecast of the epidemic outbreak.
| S-EPMC7341469 | biostudies-literature
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.
| S-EPMC6181119 | biostudies-literature
Project description:Host genetics affect both the susceptibility and response to viral infection. Searching for host genes that contribute to COVID-19, the Host Genetics Initiative (HGI) was formed to investigate the genetic factors involved in COVID-19 via genome-wide association studies (GWAS). The GWAS suffer from limited statistical power and in general, only a few genes can pass the conventional significance thresholds. This statistical limitation may be overcome by boosting weak association signals through integrating independent functional information such as molecular interactions. Additionally, the boosted results can be evaluated by various independent data for further connections to COVID-19. We present COVID-GWAB, a web-based tool to boost original GWAS signals from COVID-19 patients by taking the signals of the interactome neighbors. COVID-GWAB takes summary statistics from the COVID-19 HGI or user input data and reprioritizes candidate host genes for COVID-19 using HumanNet, a co-functional human gene network. The current version of COVID-GWAB provides the pre-processed data of releases 5, 6, and 7 of the HGI. Additionally, COVID-GWAB provides web interfaces for a summary of augmented GWAS signals, prediction evaluations by appearance frequency in COVID-19 literature, single-cell transcriptome data, and associated pathways. The web server also enables browsing the candidate gene networks.
| S-EPMC9599684 | biostudies-literature
Project description:COVID-19 disease causes serious respiratory illnesses. Therefore, accurate identification of the viral infection cycle plays a key role in designing appropriate vaccines. The risk of this disease depends on proteins that interact with human receptors. In this paper, we formulate a novel model for COVID-19 named "amino acid encoding based prediction" (AAPred). This model is accurate, classifies the various coronavirus types, and distinguishes SARS-CoV-2 from other coronaviruses. With the AAPred model, we reduce the number of features to enhance its performance by selecting the most important ones employing statistical criteria. The protein sequence of SARS-CoV-2 for understanding the viral infection cycle is analyzed. Six machine learning classifiers related to decision trees, k-nearest neighbors, random forest, support vector machine, bagging ensemble, and gradient boosting are used to evaluate the model in terms of accuracy, precision, sensitivity, and specificity. We implement the obtained results computationally and apply them to real data from the National Genomics Data Center. The experimental results report that the AAPred model reduces the features to seven of them. The average accuracy of the 10-fold cross-validation is 98.69%, precision is 98.72%, sensitivity is 96.81%, and specificity is 97.72%. The features are selected utilizing information gain and classified with random forest. The proposed model predicts the type of Coronavirus and reduces the number of extracted features. We identify that SARS-CoV-2 has similar physicochemical characteristics in some regions of SARS-CoV. Also, we report that SARS-CoV-2 has similar infection cycles and sequences in some regions of SARS CoV indicating the affectedness of vaccines on SARS-CoV-2. A comparison with deep learning shows similar results with our method.
| S-EPMC8923015 | biostudies-literature
Project description:Breast cancer is the most common invasive cancer and the second leading cause of cancer death in women. and regrettably, this rate is increasing every year. One of the aspects of all cancers, including breast cancer, is the recurrence of the disease, which causes painful consequences to the patients. Moreover, the practical application of data mining in the field of breast cancer can help to provide some necessary information and knowledge required by physicians for accurate prediction of breast cancer recurrence and better decision-making. The main objective of this study is to compare different data mining algorithms to select the most accurate model for predicting breast cancer recurrence. This study is cross-sectional and data gathering of this research performed from June 2018 to June 2019 from the official statistics of Ministry of Health and Medical Education and the Iran Cancer Research Center for patients with breast cancer who had been followed for a minimum of 5 years from February 2014 to April 2019, including 5471 independent records. After initial pre-processing in dataset and variables, seven new and conventional data mining algorithms have been applied that each one represents one kind of data mining approach. Results show that the C5.0 algorithm possibly could be a helpful tool for the prediction of breast cancer recurrence at the stage of distant recurrence and nonrecurrence, especially in the first to third years. also, LN involvement rate, Her2 value, Tumor size, free or closed tumor margin were found to be the most important features in our dataset to predict breast cancer recurrence.
| S-EPMC7561198 | biostudies-literature
Project description:Previous work using logistic regression suggests that cognitive control-related frontoparietal activation in early psychosis can predict symptomatic improvement after 1?year of coordinated specialty care with 66% accuracy. Here, we evaluated the ability of six machine learning (ML) algorithms and deep learning (DL) to predict "Improver" status (>20% improvement on Brief Psychiatric Rating Scale [BPRS] total score at 1-year follow-up vs. baseline) and continuous change in BPRS score using the same functional magnetic resonance imaging-based features (frontoparietal activations during the AX-continuous performance task) in the same sample (individuals with either schizophrenia (n?= 65, 49M/16F, mean age 20.8?years) or Type I bipolar disorder (n =?17, 9M/8F, mean age 21.6?years)). 138 healthy controls were included as a reference group. "Shallow" ML methods included Naive Bayes, support vector machine, K Star, AdaBoost, J48 decision tree, and random forest. DL included an explainable artificial intelligence (XAI) procedure for understanding results. The best overall performances (70% accuracy for the binary outcome and root mean square error = 9.47 for the continuous outcome) were achieved using DL. XAI revealed left DLPFC activation was the strongest feature used to make binary classification decisions, with a classification activation threshold (adjusted beta = .017) intermediate to the healthy control mean (adjusted beta = .15, 95% CI = -0.02 to 0.31) and patient mean (adjusted beta = -.13, 95% CI = -0.37 to 0.11). Our results suggest DL is more powerful than shallow ML methods for predicting symptomatic improvement. The left DLPFC may be a functional target for future biomarker development as its activation was particularly important for predicting improvement.
| S-EPMC7856652 | biostudies-literature
Project description:SARS-CoV-2 infection poses a worldwide public health problem affecting millions of people worldwide. There is a critical need for improvements in the noninvasive prognosis of COVID-19. We hypothesized that matrix-assisted laser desorption ionization mass spectrometry (MALDI-TOF MS) analysis combined with molecular-weight directed bottom-up proteomic analysis of plasma proteins may predict high and low risk cases of COVID-19. Patients and Methods: We used MALDI MS to analyze plasma small proteins and peptides isolated using C18 micro-columns from a cohort containing a total of 117 cases of high and low risk cases split into training (n = 88) and validation sets (n= 29). The plasma protein/peptide fingerprint obtained was used to train the algorithm before validation using a blinded test cohort. Several sample preparation, MS and data analysis parameters were optimized to achieve an overall accuracy of 85%, a sensitivity of 90%, and a specificity of 81% in the training set. In the blinded test set, this signature reached an overall accuracy of 93.1%, a sensitivity of 87.5%, and a specificity of 100%. From this signature, we identified two distinct regions corresponding to the single and doubly protonated proteins in the MALDI-TOF profile belonging to the same proteoforms. A combination of 1D SDS-PAGE and quantitative bottom-up proteomic analysis allowed the identification of intact and truncated forms of serum amyloid A-1 and A-2 proteins. We found a plasma proteomic profile that discriminates against patients with high and low risk COVID-19. Proteomic analysis of C18-fractionated plasma may have a role in the noninvasive prognosis of COVID-19. Further validation will be important to consolidate its clinical utility.
2021-06-22 | PXD021581 | Pride