Swarm Learning for decentralized and confidential clinical machine learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Warnat-Herresthal S
PROVIDER: S-EPMC8189907 | biostudies-literature | 2021 Jun
REPOSITORIES: biostudies-literature
Publications
Swarm Learning for decentralized and confidential clinical machine learning.
Warnat-Herresthal Stefanie S Schultze Hartmut H Shastry Krishnaprasad Lingadahalli KL Manamohan Sathyanarayanan S Mukherjee Saikat S Garg Vishesh V Sarveswara Ravi R Händler Kristian K Pickkers Peter P Aziz N Ahmad NA Ktena Sofia S Tran Florian F Bitzer Michael M Ossowski Stephan S Casadei Nicolas N Herr Christian C Petersheim Daniel D Behrends Uta U Kern Fabian F Fehlmann Tobias T Schommers Philipp P Lehmann Clara C Augustin Max M Rybniker Jan J Altmüller Janine J Mishra Neha N Mishra Neha N Bernardes Joana P JP Krämer Benjamin B Bonaguro Lorenzo L Schulte-Schrepping Jonas J De Domenico Elena E Siever Christian C Kraut Michael M Desai Milind M Monnet Bruno B Saridaki Maria M Siegel Charles Martin CM Drews Anna A Nuesch-Germano Melanie M Theis Heidi H Heyckendorf Jan J Schreiber Stefan S Kim-Hellmuth Sarah S Nattermann Jacob J Skowasch Dirk D Kurth Ingo I Keller Andreas A Bals Robert R Nürnberg Peter P Rieß Olaf O Rosenstiel Philip P Netea Mihai G MG Theis Fabian F Mukherjee Sach S Backes Michael M Aschenbrenner Anna C AC Ulas Thomas T Breteler Monique M B MMB Giamarellos-Bourboulis Evangelos J EJ Kox Matthijs M Becker Matthias M Cheran Sorin S Woodacre Michael S MS Goh Eng Lim EL Schultze Joachim L JL
Nature 20210526 7862
Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine<sup>1,2</sup>. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes<sup>3</sup>. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation<sup>4,5</sup>. Here, to facilitate the integration of any medical data from any data owner worldwide without violating ...[more]
Similar Datasets
Project description:Autonomous unmanned aerial vehicles (UAVs) have attracted increasing academic and industrial attention during the last decade. Using drones have broad benefits in diverse areas, such as civil and military applications, aerial photography and videography, mapping and surveying, agriculture, and disaster management. However, the recent development and innovation in the field of drone (UAV) technology have led to malicious usage of the technology, including the penetration of secure areas (such as airports) and serving terrorist attacks. Autonomous weapon systems might use drone swarms to perform more complex military tasks. Utilizing a large number of drones, simultaneously increases the risk and the reliability of the mission in terms of redundancy, survivability, scalability, and the quality of autonomous performance in a complex environment. This research suggests a new approach for drone swarm characterization and detection using RF signals analysis and various machine learning methods. While most of the existing drone detection and classification methods are typically related to a single drone classification, using supervised approaches, this research work proposes an unsupervised approach for drone swarm characterization. The proposed method utilizes the different radio frequency (RF) signatures of the drone's transmitters. Various kinds of frequency transform, such as the continuous, discrete, and wavelet scattering transform, have been applied to extract RF features from the radio frequency fingerprint, which have then been used as input for the unsupervised classifier. To reduce the input data dimension, we suggest using unsupervised approaches such as Principal component analysis (PCA), independent component analysis (ICA), uniform manifold approximation and projection (UMAP), and the t-distributed symmetric neighbor embedding (t-SNE) algorithms. The proposed clustering approach is based on common unsupervised methods, including K-means, mean shift, and X-means algorithms. The proposed approach has been evaluated using self-built and common drone swarm datasets. The results demonstrate a classification accuracy of about 95% under additive Gaussian white noise with different levels of SNR.
| S-EPMC9919887 | biostudies-literature
Project description:In recent years, comprehensive learning particle swarm optimization (CLPSO) has attracted the attention of many scholars for using in solving multimodal problems, as it is excellent in preserving the particles' diversity and thus preventing premature convergence. However, CLPSO exhibits low solution accuracy. Aiming to address this issue, we proposed a novel algorithm called LILPSO. First, this algorithm introduced a Lagrange interpolation method to perform a local search for the global best point (gbest). Second, to gain a better exemplar, one gbest, another two particle's historical best points (pbest) are chosen to perform Lagrange interpolation, then to gain a new exemplar, which replaces the CLPSO's comparison method. The numerical experiments conducted on various functions demonstrate the superiority of this algorithm, and the two methods are proven to be efficient for accelerating the convergence without leading the particle to premature convergence.
| S-EPMC4849747 | 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
Project description:BackgroundTo accelerate healthcare/genomic medicine research and facilitate quality improvement, researchers have started cross-institutional collaborations to use artificial intelligence on clinical/genomic data. However, there are real-world risks of incorrect models being submitted to the learning process, due to either unforeseen accidents or malicious intent. This may reduce the incentives for institutions to participate in the federated modeling consortium. Existing methods to deal with this "model misconduct" issue mainly focus on modifying the learning methods, and therefore are more specifically tied with the algorithm.Basic proceduresIn this paper, we aim at solving the problem in an algorithm-agnostic way by (1) designing a simulator to generate various types of model misconduct, (2) developing a framework to detect the model misconducts, and (3) providing a generalizable approach to identify model misconducts for federated learning. We considered the following three categories: Plagiarism, Fabrication, and Falsification, and then developed a detection framework with three components: Auditing, Coefficient, and Performance detectors, with greedy parameter tuning.Main findingsWe generated 10 types of misconducts from models learned on three datasets to evaluate our detection method. Our experiments showed high recall with low added computational cost. Our proposed detection method can best identify the misconduct on specific sites from any learning iteration, whereas it is more challenging to precisely detect misconducts for a specific site and at a specific iteration.Principal conclusionsWe anticipate our study can support the enhancement of the integrity and reliability of federated machine learning on genomic/healthcare data.
| S-EPMC10017272 | biostudies-literature
Project description:INTRODUCTION:Adverse event (AE) under-reporting has been a recurrent issue raised during health authorities Good Clinical Practices (GCP) inspections and audits. Moreover, safety under-reporting poses a risk to patient safety and data integrity. The current clinical Quality Assurance (QA) practices used to detect AE under-reporting rely heavily on investigator site and study audits. Yet several sponsors and institutions have had repeated findings related to safety reporting, and this has led to delays in regulatory submissions. Recent developments in data management and IT systems allow data scientists to apply techniques such as machine learning to detect AE under-reporting in an automated fashion. OBJECTIVE:In this project, we developed a predictive model that enables Roche/Genentech Quality Program Leads oversight of AE reporting at the program, study, site, and patient level. This project was part of a broader effort at Roche/Genentech Product Development Quality to apply advanced analytics to augment and complement traditional clinical QA approaches. METHOD:We used a curated data set from 104 completed Roche/Genentech sponsored clinical studies to train a machine learning model to predict the expected number of AEs. Our final model used 54 features built on patient (e.g., demographics, vitals) and study attributes (e.g., molecule class, disease area). RESULTS:In order to evaluate model performance, we tested how well it would detect simulated test cases based on data not used for model training. For relevant simulation scenarios of 25%, 50%, and 75% under-reporting on the site level, our model scored an area under the curve (AUC) of the receiver operating characteristic (ROC) curve of 0.62, 0.79, and 0.92, respectively. CONCLUSION:The model has been deployed to evaluate safety reporting performance in a set of ongoing studies in the form of a QA/dashboard cockpit available to Roche Quality Program Leads. Applicability and production performance will be assessed over the next 12-24 months in which we will develop a validation strategy to fully integrate our model into Roche QA processes.
| S-EPMC6689279 | biostudies-other
Project description:ImportanceThe development of artificial intelligence (AI)-based melanoma classifiers typically calls for large, centralized datasets, requiring hospitals to give away their patient data, which raises serious privacy concerns. To address this concern, decentralized federated learning has been proposed, where classifier development is distributed across hospitals.ObjectiveTo investigate whether a more privacy-preserving federated learning approach can achieve comparable diagnostic performance to a classical centralized (ie, single-model) and ensemble learning approach for AI-based melanoma diagnostics.Design, setting, and participantsThis multicentric, single-arm diagnostic study developed a federated model for melanoma-nevus classification using histopathological whole-slide images prospectively acquired at 6 German university hospitals between April 2021 and February 2023 and benchmarked it using both a holdout and an external test dataset. Data analysis was performed from February to April 2023.ExposuresAll whole-slide images were retrospectively analyzed by an AI-based classifier without influencing routine clinical care.Main outcomes and measuresThe area under the receiver operating characteristic curve (AUROC) served as the primary end point for evaluating the diagnostic performance. Secondary end points included balanced accuracy, sensitivity, and specificity.ResultsThe study included 1025 whole-slide images of clinically melanoma-suspicious skin lesions from 923 patients, consisting of 388 histopathologically confirmed invasive melanomas and 637 nevi. The median (range) age at diagnosis was 58 (18-95) years for the training set, 57 (18-93) years for the holdout test dataset, and 61 (18-95) years for the external test dataset; the median (range) Breslow thickness was 0.70 (0.10-34.00) mm, 0.70 (0.20-14.40) mm, and 0.80 (0.30-20.00) mm, respectively. The federated approach (0.8579; 95% CI, 0.7693-0.9299) performed significantly worse than the classical centralized approach (0.9024; 95% CI, 0.8379-0.9565) in terms of AUROC on a holdout test dataset (pairwise Wilcoxon signed-rank, P < .001) but performed significantly better (0.9126; 95% CI, 0.8810-0.9412) than the classical centralized approach (0.9045; 95% CI, 0.8701-0.9331) on an external test dataset (pairwise Wilcoxon signed-rank, P < .001). Notably, the federated approach performed significantly worse than the ensemble approach on both the holdout (0.8867; 95% CI, 0.8103-0.9481) and external test dataset (0.9227; 95% CI, 0.8941-0.9479).Conclusions and relevanceThe findings of this diagnostic study suggest that federated learning is a viable approach for the binary classification of invasive melanomas and nevi on a clinically representative distributed dataset. Federated learning can improve privacy protection in AI-based melanoma diagnostics while simultaneously promoting collaboration across institutions and countries. Moreover, it may have the potential to be extended to other image classification tasks in digital cancer histopathology and beyond.
| S-EPMC10851139 | biostudies-literature
Project description:AI is a broad concept, grouping initiatives that use a computer to perform tasks that would usually require a human to complete. AI methods are well suited to predict clinical outcomes. In practice, AI methods can be thought of as functions that learn the outcomes accompanying standardized input data to produce accurate outcome predictions when trialed with new data. Current methods for cleaning, creating, accessing, extracting, augmenting, and representing data for training AI clinical prediction models are well defined. The use of AI to predict clinical outcomes is a dynamic and rapidly evolving arena, with new methods and applications emerging. Extraction or accession of electronic health care records and combining these with patient genetic data is an area of present attention, with tremendous potential for future growth. Machine learning approaches, including decision tree methods of Random Forest and XGBoost, and deep learning techniques including deep multi-layer and recurrent neural networks, afford unique capabilities to accurately create predictions from high dimensional, multimodal data. Furthermore, AI methods are increasing our ability to accurately predict clinical outcomes that previously were difficult to model, including time-dependent and multi-class outcomes. Barriers to robust AI-based clinical outcome model deployment include changing AI product development interfaces, the specificity of regulation requirements, and limitations in ensuring model interpretability, generalizability, and adaptability over time.
| S-EPMC8786279 | biostudies-literature
Project description:BACKGROUND:Clinical narratives represent the main form of communication within health care, providing a personalized account of patient history and assessments, and offering rich information for clinical decision making. Natural language processing (NLP) has repeatedly demonstrated its feasibility to unlock evidence buried in clinical narratives. Machine learning can facilitate rapid development of NLP tools by leveraging large amounts of text data. OBJECTIVE:The main aim of this study was to provide systematic evidence on the properties of text data used to train machine learning approaches to clinical NLP. We also investigated the types of NLP tasks that have been supported by machine learning and how they can be applied in clinical practice. METHODS:Our methodology was based on the guidelines for performing systematic reviews. In August 2018, we used PubMed, a multifaceted interface, to perform a literature search against MEDLINE. We identified 110 relevant studies and extracted information about text data used to support machine learning, NLP tasks supported, and their clinical applications. The data properties considered included their size, provenance, collection methods, annotation, and any relevant statistics. RESULTS:The majority of datasets used to train machine learning models included only hundreds or thousands of documents. Only 10 studies used tens of thousands of documents, with a handful of studies utilizing more. Relatively small datasets were utilized for training even when much larger datasets were available. The main reason for such poor data utilization is the annotation bottleneck faced by supervised machine learning algorithms. Active learning was explored to iteratively sample a subset of data for manual annotation as a strategy for minimizing the annotation effort while maximizing the predictive performance of the model. Supervised learning was successfully used where clinical codes integrated with free-text notes into electronic health records were utilized as class labels. Similarly, distant supervision was used to utilize an existing knowledge base to automatically annotate raw text. Where manual annotation was unavoidable, crowdsourcing was explored, but it remains unsuitable because of the sensitive nature of data considered. Besides the small volume, training data were typically sourced from a small number of institutions, thus offering no hard evidence about the transferability of machine learning models. The majority of studies focused on text classification. Most commonly, the classification results were used to support phenotyping, prognosis, care improvement, resource management, and surveillance. CONCLUSIONS:We identified the data annotation bottleneck as one of the key obstacles to machine learning approaches in clinical NLP. Active learning and distant supervision were explored as a way of saving the annotation efforts. Future research in this field would benefit from alternatives such as data augmentation and transfer learning, or unsupervised learning, which do not require data annotation.
| S-EPMC7157505 | biostudies-literature
Project description:BackgroundTraditional clinical assessments often lack individualization, relying on standardized procedures that may not accommodate the diverse needs of patients, especially in early stages where personalized diagnosis could offer significant benefits. We aim to provide a machine-learning framework that addresses the individualized feature addition problem and enhances diagnostic accuracy for clinical assessments.MethodsIndividualized Clinical Assessment Recommendation System (iCARE) employs locally weighted logistic regression and Shapley Additive Explanations (SHAP) value analysis to tailor feature selection to individual patient characteristics. Evaluations were conducted on synthetic and real-world datasets, including early-stage diabetes risk prediction and heart failure clinical records from the UCI Machine Learning Repository. We compared the performance of iCARE with a Global approach using statistical analysis on accuracy and area under the ROC curve (AUC) to select the best additional features.FindingsThe iCARE framework enhances predictive accuracy and AUC metrics when additional features exhibit distinct predictive capabilities, as evidenced by synthetic datasets 1-3 and the early diabetes dataset. Specifically, in synthetic dataset 1, iCARE achieved an accuracy of 0·999 and an AUC of 1·000, outperforming the Global approach with an accuracy of 0·689 and an AUC of 0·639. In the early diabetes dataset, iCARE shows improvements of 1·5-3·5% in accuracy and AUC across different numbers of initial features. Conversely, in synthetic datasets 4-5 and the heart failure dataset, where features lack discernible predictive distinctions, iCARE shows no significant advantage over global approaches on accuracy and AUC metrics.InterpretationiCARE provides personalized feature recommendations that enhance diagnostic accuracy in scenarios where individualized approaches are critical, improving the precision and effectiveness of medical diagnoses.FundingThis work was supported by startup funding from the Department of Psychology at the University of Kansas provided to A.A., and the R01MH125740 award from NIH partially supported J.M.G.'s work.
| S-EPMC11302612 | biostudies-literature
Project description:PurposeRecently a realistic simulator of patient seizure diaries was developed that can reproduce effects seen in randomized clinical trials (RCTs). RCTs suffer from high costs and statistical inefficiencies. Using realistic simulation and machine learning this study aimed to identify a more statistically efficient outcome metric.MethodsFive candidate deep learning architectures with 54 permutations of hyperparameters were compared to the traditional standard, median percent change (MPC). Each were also tested for type 1 error. All models had similar outcomes, with appropriate low levels of type 1 error.ResultsThe simplest model was equivalent to a logistic regression of a histogram of individual percentage changes in seizure rate, requiring 21-22% less patients to discriminate drug from placebo at 90% power. This model was referred to as LPC.ConclusionFuture studies to validate LPC may enable faster, cheaper and more efficient clinical trials.
| S-EPMC8435025 | biostudies-literature