An adversarial training framework for mitigating algorithmic biases in clinical machine learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Yang J
PROVIDER: S-EPMC10050816 | biostudies-literature | 2023 Mar
REPOSITORIES: biostudies-literature
Publications
An adversarial training framework for mitigating algorithmic biases in clinical machine learning.
Yang Jenny J Soltan Andrew A S AAS Eyre David W DW Yang Yang Y Clifton David A DA
NPJ digital medicine 20230329 1
Machine learning is becoming increasingly prominent in healthcare. Although its benefits are clear, growing attention is being given to how these tools may exacerbate existing biases and disparities. In this study, we introduce an adversarial training framework that is capable of mitigating biases that may have been acquired through data collection. We demonstrate this proposed framework on the real-world task of rapidly predicting COVID-19, and focus on mitigating site-specific (hospital) and d ...[more]
Similar Datasets
Project description:As models based on machine learning continue to be developed for healthcare applications, greater effort is needed to ensure that these technologies do not reflect or exacerbate any unwanted or discriminatory biases that may be present in the data. Here we introduce a reinforcement learning framework capable of mitigating biases that may have been acquired during data collection. In particular, we evaluated our model for the task of rapidly predicting COVID-19 for patients presenting to hospital emergency departments and aimed to mitigate any site (hospital)-specific and ethnicity-based biases present in the data. Using a specialized reward function and training procedure, we show that our method achieves clinically effective screening performances, while significantly improving outcome fairness compared with current benchmarks and state-of-the-art machine learning methods. We performed external validation across three independent hospitals, and additionally tested our method on a patient intensive care unit discharge status task, demonstrating model generalizability.
| S-EPMC10442224 | biostudies-literature
Project description:BackgroundOpioid use disorder (OUD) is a critical public health crisis in the United States, affecting >5.5 million Americans in 2021. Machine learning has been used to predict patient risk of incident OUD. However, little is known about the fairness and bias of these predictive models.ObjectiveThe aims of this study are two-fold: (1) to develop a machine learning bias mitigation algorithm for sociodemographic features and (2) to develop a fairness-aware weighted majority voting (WMV) classifier for OUD prediction.MethodsWe used the 2020 National Survey on Drug and Health data to develop a neural network (NN) model using stochastic gradient descent (SGD; NN-SGD) and an NN model using Adam (NN-Adam) optimizers and evaluated sociodemographic bias by comparing the area under the curve values. A bias mitigation algorithm, based on equality of odds, was implemented to minimize disparities in specificity and recall. Finally, a WMV classifier was developed for fairness-aware prediction of OUD. To further analyze bias detection and mitigation, we did a 1-N matching of OUD to non-OUD cases, controlling for socioeconomic variables, and evaluated the performance of the proposed bias mitigation algorithm and WMV classifier.ResultsOur bias mitigation algorithm substantially reduced bias with NN-SGD, by 21.66% for sex, 1.48% for race, and 21.04% for income, and with NN-Adam by 16.96% for sex, 8.87% for marital status, 8.45% for working condition, and 41.62% for race. The fairness-aware WMV classifier achieved a recall of 85.37% and 92.68% and an accuracy of 58.85% and 90.21% using NN-SGD and NN-Adam, respectively. The results after matching also indicated remarkable bias reduction with NN-SGD and NN-Adam, respectively, as follows: sex (0.14% vs 0.97%), marital status (12.95% vs 10.33%), working condition (14.79% vs 15.33%), race (60.13% vs 41.71%), and income (0.35% vs 2.21%). Moreover, the fairness-aware WMV classifier achieved high performance with a recall of 100% and 85.37% and an accuracy of 73.20% and 89.38% using NN-SGD and NN-Adam, respectively.ConclusionsThe application of the proposed bias mitigation algorithm shows promise in reducing sociodemographic bias, with the WMV classifier confirming bias reduction and high performance in OUD prediction.
| S-EPMC11372321 | biostudies-literature
Project description:We show how complexity theory can be introduced in machine learning to help bring together apparently disparate areas of current research. We show that this model-driven approach may require less training data and can potentially be more generalizable as it shows greater resilience to random attacks. In an algorithmic space the order of its element is given by its algorithmic probability, which arises naturally from computable processes. We investigate the shape of a discrete algorithmic space when performing regression or classification using a loss function parametrized by algorithmic complexity, demonstrating that the property of differentiation is not required to achieve results similar to those obtained using differentiable programming approaches such as deep learning. In doing so we use examples which enable the two approaches to be compared (small, given the computational power required for estimations of algorithmic complexity). We find and report that 1) machine learning can successfully be performed on a non-smooth surface using algorithmic complexity; 2) that solutions can be found using an algorithmic-probability classifier, establishing a bridge between a fundamentally discrete theory of computability and a fundamentally continuous mathematical theory of optimization methods; 3) a formulation of an algorithmically directed search technique in non-smooth manifolds can be defined and conducted; 4) exploitation techniques and numerical methods for algorithmic search to navigate these discrete non-differentiable spaces can be performed; in application of the (a) identification of generative rules from data observations; (b) solutions to image classification problems more resilient against pixel attacks compared to neural networks; (c) identification of equation parameters from a small data-set in the presence of noise in continuous ODE system problem, (d) classification of Boolean NK networks by (1) network topology, (2) underlying Boolean function, and (3) number of incoming edges.
| S-EPMC7944352 | biostudies-literature
Project description:Transcriptional enhancers are non-coding segments of DNA that play a central role in the spatiotemporal regulation of gene expression programs. However, systematically and precisely predicting enhancers remain a major challenge. Although existing methods have achieved some success in enhancer prediction, they still suffer from many issues. We developed a deep learning-based algorithmic framework named PEDLA (https://github.com/wenjiegroup/PEDLA), which can directly learn an enhancer predictor from massively heterogeneous data and generalize in ways that are mostly consistent across various cell types/tissues. We first trained PEDLA with 1,114-dimensional heterogeneous features in H1 cells, and demonstrated that PEDLA framework integrates diverse heterogeneous features and gives state-of-the-art performance relative to five existing methods for enhancer prediction. We further extended PEDLA to iteratively learn from 22 training cell types/tissues. Our results showed that PEDLA manifested superior performance consistency in both training and independent test sets. On average, PEDLA achieved 95.0% accuracy and a 96.8% geometric mean (GM) of sensitivity and specificity across 22 training cell types/tissues, as well as 95.7% accuracy and a 96.8% GM across 20 independent test cell types/tissues. Together, our work illustrates the power of harnessing state-of-the-art deep learning techniques to consistently identify regulatory elements at a genome-wide scale from massively heterogeneous data across diverse cell types/tissues.
| S-EPMC4916453 | biostudies-literature
Project description:ImportancePredictive models using machine learning techniques have potential to improve early detection and management of Alzheimer disease (AD). However, these models potentially have biases and may perpetuate or exacerbate existing disparities.ObjectiveTo characterize the algorithmic fairness of longitudinal prediction models for AD progression.Design, setting, and participantsThis prognostic study investigated the algorithmic fairness of logistic regression, support vector machines, and recurrent neural networks for predicting progression to mild cognitive impairment (MCI) and AD using data from participants in the Alzheimer Disease Neuroimaging Initiative evaluated at 57 sites in the US and Canada. Participants aged 54 to 91 years who contributed data on at least 2 visits between September 2005 and May 2017 were included. Data were analyzed in October 2022.ExposuresFairness was quantified across sex, ethnicity, and race groups. Neuropsychological test scores, anatomical features from T1 magnetic resonance imaging, measures extracted from positron emission tomography, and cerebrospinal fluid biomarkers were included as predictors.Main outcomes and measuresOutcome measures quantified fairness of prediction models (logistic regression [LR], support vector machine [SVM], and recurrent neural network [RNN] models), including equal opportunity, equalized odds, and demographic parity. Specifically, if the model exhibited equal sensitivity for all groups, it aligned with the principle of equal opportunity, indicating fairness in predictive performance.ResultsA total of 1730 participants in the cohort (mean [SD] age, 73.81 [6.92] years; 776 females [44.9%]; 69 Hispanic [4.0%] and 1661 non-Hispanic [96.0%]; 29 Asian [1.7%], 77 Black [4.5%], 1599 White [92.4%], and 25 other race [1.4%]) were included. Sensitivity for predicting progression to MCI and AD was lower for Hispanic participants compared with non-Hispanic participants; the difference (SD) in true positive rate ranged from 20.9% (5.5%) for the RNN model to 27.8% (9.8%) for the SVM model in MCI and 24.1% (5.4%) for the RNN model to 48.2% (17.3%) for the LR model in AD. Sensitivity was similarly lower for Black and Asian participants compared with non-Hispanic White participants; for example, the difference (SD) in AD true positive rate was 14.5% (51.6%) in the LR model, 12.3% (35.1%) in the SVM model, and 28.4% (16.8%) in the RNN model for Black vs White participants, and the difference (SD) in MCI true positive rate was 25.6% (13.1%) in the LR model, 24.3% (13.1%) in the SVM model, and 6.8% (18.7%) in the RNN model for Asian vs White participants. Models generally satisfied metrics of fairness with respect to sex, with no significant differences by group, except for cognitively normal (CN)-MCI and MCI-AD transitions (eg, an absolute increase [SD] in the true positive rate of CN-MCI transitions of 10.3% [27.8%] for the LR model).Conclusions and relevanceIn this study, models were accurate in aggregate but failed to satisfy fairness metrics. These findings suggest that fairness should be considered in the development and use of machine learning models for AD progression.
| S-EPMC10630899 | biostudies-literature
Project description:Metabolomics provides a direct functional readout of a tumor’s physiology. Yet, it is lagging behind other omics technologies in facilitating disease monitoring and prognostication. This stems partly from the scarcity of large-scale metabolomic studies, but also the analytical complexities of detecting diverse metabolites with varying physicochemical properties and concentrations. To address this, we developed a machine learning framework using both tumor tissue and cell line samples across multiple cancer types that allows prediction of metabolomics from gene expression data. Two different model types were selected and trained for tissues and cell lines with their generalization capacity validated on independent cohorts, accurately predicting as high as 70-80% of tested metabolites. This work offers a scalable and efficient machine learning pipeline to determine metabolic from transcriptomic signatures, opening avenues to reconstruct and study the metabolic landscape of samples across novel and existing datasets lacking direct metabolomics measurements.
2025-09-30 | GSE299429 | GEO
Project description:Tears within the stabilizing muscles of the shoulder, known as the rotator cuff (RC), are the most common cause of shoulder pain-often presenting in older patients and requiring expensive advanced imaging for diagnosis. Despite the high prevalence of RC tears within the elderly population, there is no previously published work examining shoulder kinematics using markerless motion capture in the context of shoulder injury. Here we show that a simple string pulling behavior task, where subjects pull a string using hand-over-hand motions, provides a reliable readout of shoulder mobility across animals and humans. We find that both mice and humans with RC tears exhibit decreased movement amplitude, prolonged movement time, and quantitative changes in waveform shape during string pulling task performance. In rodents, we further note the degradation of low dimensional, temporally coordinated movements after injury. Furthermore, a logistic regression model built on our biomarker ensemble succeeds in classifying human patients as having a RC tear with > 90% accuracy. Our results demonstrate how a combined framework bridging animal models, motion capture, convolutional neural networks, and algorithmic assessment of movement quality enables future research into the development of smartphone-based, at-home diagnostic tests for shoulder injury.
| S-EPMC10652003 | biostudies-literature
Project description:Adversarial machine learning is a recent area of study that explores both adversarial attack strategy and detection systems of adversarial attacks, which are inputs specially crafted to outwit the classification of detection systems or disrupt the training process of detection systems. In this research, we performed two adversarial attack scenarios, we used a Generative Adversarial Network (GAN) to generate synthetic intrusion traffic to test the influence of these attacks on the accuracy of machine learning-based Intrusion Detection Systems(IDSs). We conducted two experiments on adversarial attacks including poisoning and evasion attacks on two different types of machine learning models: Decision Tree and Logistic Regression. The performance of implemented adversarial attack scenarios was evaluated using the CICIDS2017 dataset. Also, it was based on a comparison of the accuracy of machine learning-based IDS before and after attacks. The results show that the proposed evasion attacks reduced the testing accuracy of both network intrusion detection systems models (NIDS). That illustrates our evasion attack scenario negatively affected the accuracy of machine learning-based network intrusion detection systems, whereas the decision tree model was more affected than logistic regression. Furthermore, our poisoning attack scenario disrupted the training process of machine learning-based NIDS, whereas the logistic regression model was more affected than the decision tree.
| S-EPMC9565394 | biostudies-literature
Project description:In recent years, the spread of social networking services (SNS) has made it easier to connect with people who have similar opinions. Accordingly, similar opinions are shared within a group, while the frequency of exposure to different opinions tends to decrease. As a result, the polarization of opinion among groups is more likely to occur. Some studies have been conducted to identify the conditions under which opinion polarization occurs by simulating opinion dynamics, but specific methods for mitigating it have been poorly understood. Recently, it was found that even a few artificial perturbations inspired by the adversarial attack reverse the result in voter models, where a minority opinion becomes dominant through these perturbations. In this study, we conducted numerical simulations to determine whether it is possible to mitigate opinion polarization by adding such perturbations to the network in opinion dynamics models. The results show that opinion polarization can be mitigated by strategically generating perturbations to the weights of network links, and the effect increases as the perturbation strength parameter increases. Moreover, our analysis reveals that the effectiveness of this polarization mitigating method is enhanced in larger networks. Our results propose an effective way to prevent polarization of opinion in social networks.
| S-EPMC11911447 | biostudies-literature
Project description:BackgroundMachine learning involves strategies and algorithms that may assist bioinformatics analyses in terms of data mining and knowledge discovery. In several applications, viz. in Life Sciences, it is often more important to understand how a prediction was obtained rather than knowing what prediction was made. To this end so-called interpretable machine learning has been recently advocated. In this study, we implemented an interpretable machine learning package based on the rough set theory. An important aim of our work was provision of statistical properties of the models and their components.ResultsWe present the R.ROSETTA package, which is an R wrapper of ROSETTA framework. The original ROSETTA functions have been improved and adapted to the R programming environment. The package allows for building and analyzing non-linear interpretable machine learning models. R.ROSETTA gathers combinatorial statistics via rule-based modelling for accessible and transparent results, well-suited for adoption within the greater scientific community. The package also provides statistics and visualization tools that facilitate minimization of analysis bias and noise. The R.ROSETTA package is freely available at https://github.com/komorowskilab/R.ROSETTA . To illustrate the usage of the package, we applied it to a transcriptome dataset from an autism case-control study. Our tool provided hypotheses for potential co-predictive mechanisms among features that discerned phenotype classes. These co-predictors represented neurodevelopmental and autism-related genes.ConclusionsR.ROSETTA provides new insights for interpretable machine learning analyses and knowledge-based systems. We demonstrated that our package facilitated detection of dependencies for autism-related genes. Although the sample application of R.ROSETTA illustrates transcriptome data analysis, the package can be used to analyze any data organized in decision tables.
| S-EPMC7937228 | biostudies-literature