Data leakage inflates prediction performance in connectome-based machine learning models.
Ontology highlight
ABSTRACT:
SUBMITTER: Rosenblatt M
PROVIDER: S-EPMC10901797 | biostudies-literature | 2024 Feb
REPOSITORIES: biostudies-literature
Publications
Data leakage inflates prediction performance in connectome-based machine learning models.
Rosenblatt Matthew M Tejavibulya Link L Jiang Rongtao R Noble Stephanie S Scheinost Dustin D
Nature communications 20240228 1
Predictive modeling is a central technique in neuroimaging to identify brain-behavior relationships and test their generalizability to unseen data. However, data leakage undermines the validity of predictive models by breaching the separation between training and test data. Leakage is always an incorrect practice but still pervasive in machine learning. Understanding its effects on neuroimaging predictive models can inform how leakage affects existing literature. Here, we investigate the effects ...[more]
Similar Datasets
Project description:Predictive modeling has now become a central technique in neuroimaging to identify complex brain-behavior relationships and test their generalizability to unseen data. However, data leakage, which unintentionally breaches the separation between data used to train and test the model, undermines the validity of predictive models. Previous literature suggests that leakage is generally pervasive in machine learning, but few studies have empirically evaluated the effects of leakage in neuroimaging data. Although leakage is always an incorrect practice, understanding the effects of leakage on neuroimaging predictive models provides insight into the extent to which leakage may affect the literature. Here, we investigated the effects of leakage on machine learning models in two common neuroimaging modalities, functional and structural connectomes. Using over 400 different pipelines spanning four large datasets and three phenotypes, we evaluated five forms of leakage fitting into three broad categories: feature selection, covariate correction, and lack of independence between subjects. As expected, leakage via feature selection and repeated subjects drastically inflated prediction performance. Notably, other forms of leakage had only minor effects (e.g., leaky site correction) or even decreased prediction performance (e.g., leaky covariate regression). In some cases, leakage affected not only prediction performance, but also model coefficients, and thus neurobiological interpretations. Finally, we found that predictive models using small datasets were more sensitive to leakage. Overall, our results illustrate the variable effects of leakage on prediction pipelines and underscore the importance of avoiding data leakage to improve the validity and reproducibility of predictive modeling.
| S-EPMC10793416 | biostudies-literature
Project description:Neuroimaging-based predictive models continue to improve in performance, yet a widely overlooked aspect of these models is "trustworthiness," or robustness to data manipulations. High trustworthiness is imperative for researchers to have confidence in their findings and interpretations. In this work, we used functional connectomes to explore how minor data manipulations influence machine learning predictions. These manipulations included a method to falsely enhance prediction performance and adversarial noise attacks designed to degrade performance. Although these data manipulations drastically changed model performance, the original and manipulated data were extremely similar (r = 0.99) and did not affect other downstream analysis. Essentially, connectome data could be inconspicuously modified to achieve any desired prediction performance. Overall, our enhancement attacks and evaluation of existing adversarial noise attacks in connectome-based models highlight the need for counter-measures that improve the trustworthiness to preserve the integrity of academic research and any potential translational applications.
| S-EPMC10382940 | biostudies-literature
Project description:Transcriptome prediction methods such as PrediXcan and FUSION have become popular in complex trait mapping. Most transcriptome prediction models have been trained in European populations using methods that make parametric linear assumptions like the elastic net (EN). To potentially further optimize imputation performance of gene expression across global populations, we built transcriptome prediction models using both linear and non-linear machine learning (ML) algorithms and evaluated their performance in comparison to EN. We trained models using genotype and blood monocyte transcriptome data from the Multi-Ethnic Study of Atherosclerosis (MESA) comprising individuals of African, Hispanic, and European ancestries and tested them using genotype and whole-blood transcriptome data from the Modeling the Epidemiology Transition Study (METS) comprising individuals of African ancestries. We show that the prediction performance is highest when the training and the testing population share similar ancestries regardless of the prediction algorithm used. While EN generally outperformed random forest (RF), support vector regression (SVR), and K nearest neighbor (KNN), we found that RF outperformed EN for some genes, particularly between disparate ancestries, suggesting potential robustness and reduced variability of RF imputation performance across global populations. When applied to a high-density lipoprotein (HDL) phenotype, we show including RF prediction models in PrediXcan revealed potential gene associations missed by EN models. Therefore, by integrating other ML modeling into PrediXcan and diversifying our training populations to include more global ancestries, we may uncover new genes associated with complex traits.
| S-EPMC8087249 | biostudies-literature
Project description:The objective of this study is to evaluate machine learning algorithms aimed at predicting surgical treatment outcomes in groups of patients with temporal lobe epilepsy (TLE) using only the structural brain connectome. Specifically, the brain connectome is reconstructed using white matter fiber tracts from presurgical diffusion tensor imaging. To achieve our objective, a two-stage connectome-based prediction framework is developed that gradually selects a small number of abnormal network connections that contribute to the surgical treatment outcome, and in each stage a linear kernel operation is used to further improve the accuracy of the learned classifier. Using a 10-fold cross validation strategy, the first stage in the connectome-based framework is able to separate patients with TLE from normal controls with 80% accuracy, and second stage in the connectome-based framework is able to correctly predict the surgical treatment outcome of patients with TLE with 70% accuracy. Compared to existing state-of-the-art methods that use VBM data, the proposed two-stage connectome-based prediction framework is a suitable alternative with comparable prediction performance. Our results additionally show that machine learning algorithms that exclusively use structural connectome data can predict treatment outcomes in epilepsy with similar accuracy compared with "expert-based" clinical decision. In summary, using the unprecedented information provided in the brain connectome, machine learning algorithms may uncover pathological changes in brain network organization and improve outcome forecasting in the context of epilepsy.
| S-EPMC4701213 | biostudies-literature
Project description:BackgroundAccidental hypothermia is a critical condition with high risks of fatal arrhythmia, multiple organ failure, and mortality; however, there is no established model to predict the mortality. The present study aimed to develop and validate machine learning-based models for predicting in-hospital mortality using easily available data at hospital admission among the patients with accidental hypothermia.MethodThis study was secondary analysis of multi-center retrospective cohort study (J-point registry) including patients with accidental hypothermia. Adult patients with body temperature 35.0 °C or less at emergency department were included. Prediction models for in-hospital mortality using machine learning (lasso, random forest, and gradient boosting tree) were made in development cohort from six hospitals, and the predictive performance were assessed in validation cohort from other six hospitals. As a reference, we compared the SOFA score and 5A score.ResultsWe included total 532 patients in the development cohort [N = 288, six hospitals, in-hospital mortality: 22.0% (64/288)], and the validation cohort [N = 244, six hospitals, in-hospital mortality 27.0% (66/244)]. The C-statistics [95% CI] of the models in validation cohorts were as follows: lasso 0.784 [0.717-0.851] , random forest 0.794[0.735-0.853], gradient boosting tree 0.780 [0.714-0.847], SOFA 0.787 [0.722-0.851], and 5A score 0.750[0.681-0.820]. The calibration plot showed that these models were well calibrated to observed in-hospital mortality. Decision curve analysis indicated that these models obtained clinical net-benefit.ConclusionThis multi-center retrospective cohort study indicated that machine learning-based prediction models could accurately predict in-hospital mortality in validation cohort among the accidental hypothermia patients. These models might be able to support physicians and patient's decision-making. However, the applicability to clinical settings, and the actual clinical utility is still unclear; thus, further prospective study is warranted to evaluate the clinical usefulness.
| S-EPMC7797142 | biostudies-literature
Project description:BackgroundMachine learning algorithms are currently used in a wide array of clinical domains to produce models that can predict clinical risk events. Most models are developed and evaluated with retrospective data, very few are evaluated in a clinical workflow, and even fewer report performances in different hospitals. In this study, we provide detailed evaluations of clinical risk prediction models in live clinical workflows for three different use cases in three different hospitals.ObjectiveThe main objective of this study was to evaluate clinical risk prediction models in live clinical workflows and compare their performance in these setting with their performance when using retrospective data. We also aimed at generalizing the results by applying our investigation to three different use cases in three different hospitals.MethodsWe trained clinical risk prediction models for three use cases (ie, delirium, sepsis, and acute kidney injury) in three different hospitals with retrospective data. We used machine learning and, specifically, deep learning to train models that were based on the Transformer model. The models were trained using a calibration tool that is common for all hospitals and use cases. The models had a common design but were calibrated using each hospital's specific data. The models were deployed in these three hospitals and used in daily clinical practice. The predictions made by these models were logged and correlated with the diagnosis at discharge. We compared their performance with evaluations on retrospective data and conducted cross-hospital evaluations.ResultsThe performance of the prediction models with data from live clinical workflows was similar to the performance with retrospective data. The average value of the area under the receiver operating characteristic curve (AUROC) decreased slightly by 0.6 percentage points (from 94.8% to 94.2% at discharge). The cross-hospital evaluations exhibited severely reduced performance: the average AUROC decreased by 8 percentage points (from 94.2% to 86.3% at discharge), which indicates the importance of model calibration with data from the deployment hospital.ConclusionsCalibrating the prediction model with data from different deployment hospitals led to good performance in live settings. The performance degradation in the cross-hospital evaluation identified limitations in developing a generic model for different hospitals. Designing a generic process for model development to generate specialized prediction models for each hospital guarantees model performance in different hospitals.
| S-EPMC9214618 | biostudies-literature
Project description:The goal of precision oncology is to tailor treatment for patients individually using the genomic profile of their tumors. Pharmacogenomics datasets such as cancer cell lines are among the most valuable resources for drug sensitivity prediction, a crucial task of precision oncology. Machine learning methods have been employed to predict drug sensitivity based on the multiple omics data available for large panels of cancer cell lines. However, there are no comprehensive guidelines on how to properly train and validate such machine learning models for drug sensitivity prediction. In this paper, we introduce a set of guidelines for different aspects of training gene expression-based predictors using cell line datasets. These guidelines provide extensive analysis of the generalization of drug sensitivity predictors and challenge many current practices in the community including the choice of training dataset and measure of drug sensitivity. The application of these guidelines in future studies will enable the development of more robust preclinical biomarkers.
| S-EPMC8575017 | biostudies-literature
Project description:BackgroundProcedure-specific complications can have devastating consequences. Machine learning-based tools have the potential to outperform traditional statistical modeling in predicting their risk and guiding decision-making. We sought to develop and compare deep neural network (NN) models, a type of machine learning, to logistic regression (LR) for predicting anastomotic leak after colectomy, bile leak after hepatectomy, and pancreatic fistula after pancreaticoduodenectomy (PD).MethodsThe colectomy, hepatectomy, and PD National Surgical Quality Improvement Program (NSQIP) databases were analyzed. Each dataset was split into training, validation, and testing sets in a 60/20/20 ratio, with fivefold cross-validation. Models were created using NN and LR for each outcome. Models were evaluated primarily with area under the receiver operating characteristic curve (AUROC).ResultsA total of 197,488 patients were included for colectomy, 25,403 for hepatectomy, and 23,333 for PD. For anastomotic leak, AUROC for NN was 0.676 (95% 0.666-0.687), compared with 0.633 (95% CI 0.620-0.647) for LR. For bile leak, AUROC for NN was 0.750 (95% CI 0.739-0.761), compared with 0.722 (95% CI 0.698-0.746) for LR. For pancreatic fistula, AUROC for NN was 0.746 (95% CI 0.733-0.760), compared with 0.713 (95% CI 0.703-0.723) for LR. Variables related to intra-operative information, such as surgical approach, biliary reconstruction, and pancreatic gland texture were highly important for model predictions.DiscussionMachine learning showed a marginal advantage over traditional statistical techniques in predicting procedure-specific outcomes. However, models that included intra-operative information performed better than those that did not, suggesting that NSQIP procedure-targeted datasets may be strengthened by including relevant intra-operative information.
| S-EPMC9444966 | biostudies-literature
Project description:BackgroundData drift can negatively impact the performance of machine learning algorithms (MLAs) that were trained on historical data. As such, MLAs should be continuously monitored and tuned to overcome the systematic changes that occur in the distribution of data. In this paper, we study the extent of data drift and provide insights about its characteristics for sepsis onset prediction. This study will help elucidate the nature of data drift for prediction of sepsis and similar diseases. This may aid with the development of more effective patient monitoring systems that can stratify risk for dynamic disease states in hospitals.MethodsWe devise a series of simulations that measure the effects of data drift in patients with sepsis. We simulate multiple scenarios in which data drift may occur, namely the change in the distribution of the predictor variables (covariate shift), the change in the statistical relationship between the predictors and the target (concept shift), and the occurrence of a major healthcare event (major event) such as the COVID-19 pandemic. We measure the impact of data drift on model performances, identify the circumstances that necessitate model retraining, and compare the effects of different retraining methodologies and model architecture on the outcomes. We present the results for two different MLAs, eXtreme Gradient Boosting (XGB) and Recurrent Neural Network (RNN).ResultsOur results show that the properly retrained XGB models outperform the baseline models in all simulation scenarios, hence signifying the existence of data drift. In the major event scenario, the area under the receiver operating characteristic curve (AUROC) at the end of the simulation period is 0.811 for the baseline XGB model and 0.868 for the retrained XGB model. In the covariate shift scenario, the AUROC at the end of the simulation period for the baseline and retrained XGB models is 0.853 and 0.874 respectively. In the concept shift scenario and under the mixed labeling method, the retrained XGB models perform worse than the baseline model for most simulation steps. However, under the full relabeling method, the AUROC at the end of the simulation period for the baseline and retrained XGB models is 0.852 and 0.877 respectively. The results for the RNN models were mixed, suggesting that retraining based on a fixed network architecture may be inadequate for an RNN. We also present the results in the form of other performance metrics such as the ratio of observed to expected probabilities (calibration) and the normalized rate of positive predictive values (PPV) by prevalence, referred to as lift, at a sensitivity of 0.8.ConclusionOur simulations reveal that retraining periods of a couple of months or using several thousand patients are likely to be adequate to monitor machine learning models that predict sepsis. This indicates that a machine learning system for sepsis prediction will probably need less infrastructure for performance monitoring and retraining compared to other applications in which data drift is more frequent and continuous. Our results also show that in the event of a concept shift, a full overhaul of the sepsis prediction model may be necessary because it indicates a discrete change in the definition of sepsis labels, and mixing the labels for the sake of incremental training may not produce the desired results.
| S-EPMC9196120 | biostudies-literature
Project description:Gene expression profiles were generated from 199 primary breast cancer patients. Samples 1-176 were used in another study, GEO Series GSE22820, and form the training data set in this study. Sample numbers 200-222 form a validation set. This data is used to model a machine learning classifier for Estrogen Receptor Status. RNA was isolated from 199 primary breast cancer patients. A machine learning classifier was built to predict ER status using only three gene features.
2013-01-01 | E-GEOD-29210 | biostudies-arrayexpress