Semi-supervised active transfer learning for fetal ECG arrhythmia detection

Mohebbian, Mohammad Reza; Marateb, Hamid Reza; Wahid, Khan A.

doi:10.1016/j.cmpbup.2023.100096

Cited by 18 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In 1967, Cover and Hart designed a simple supervised machine learning approach to categorize patterns [ 30 ]. The algorithm operates by sorting query data points according to the votes received from the k-nearest data points.…”

Section: Methodsmentioning

confidence: 99%

Machine Learning-Based Virtual Screening and Molecular Simulation Approaches Identified Novel Potential Inhibitors for Cancer Therapy

et al. 2023

View full text Add to dashboard Cite

Cyclin-dependent kinase 2 (CDK2) is a promising target for cancer treatment, developing new effective CDK2 inhibitors is of great significance in anticancer therapy. The involvement of CDK2 in tumorigenesis has been debated, but recent evidence suggests that specifically inhibiting CDK2 could be beneficial in treating certain tumors. This approach remains attractive in the development of anticancer drugs. Several small-molecule inhibitors targeting CDK2 have reached clinical trials, but a selective inhibitor for CDK2 is yet to be discovered. In this study, we conducted machine learning-based drug designing to search for a drug candidate for CDK2. Machine learning models, including k-NN, SVM, RF, and GNB, were created to detect active and inactive inhibitors for a CDK2 drug target. The models were assessed using 10-fold cross-validation to ensure their accuracy and reliability. These methods are highly suitable for classifying compounds as either active or inactive through the virtual screening of extensive compound libraries. Subsequently, machine learning techniques were employed to analyze the test dataset obtained from the zinc database. A total of 25 compounds with 98% accuracy were predicted as active against CDK2. These compounds were docked into CDK2’s active site. Finally, three compounds were selected based on good docking score, and, along with a reference compound, underwent MD simulation. The Gaussian naïve Bayes model yielded superior results compared to other models. The top three hits exhibited enhanced stability and compactness compared to the reference compound. In conclusion, our study provides valuable insights for identifying and refining lead compounds as CDK2 inhibitors.

show abstract

Section: Methodsmentioning

confidence: 99%

Machine Learning-Based Virtual Screening and Molecular Simulation Approaches Identified Novel Potential Inhibitors for Cancer Therapy

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Фреймворк використовує часткові анотації, навчання з частковим залученням вчителя і оцінюванням розмічування, що дозволило значно скоротити час, необхідний для точної сегментації великих наборів гістологічних даних. А у роботі [8] активне навчання було використано для розмічування набору даних призначеного для виявлення аритмії на електрокардіограмі плода .…”

Section: рисунок 1 -цикл активного навчанняunclassified

Використання Підходів Активного Навчання Під Час Побудови Моделей Машинного Навчання

Yemelyanenko

2022

APAIT

View full text Add to dashboard Cite

Дніпровський національний університет імені Олеся Гончара ВИКОРИСТАННЯ ПІДХОДІВ АКТИВНОГО НАВЧАННЯ ПІД ЧАС ПОБУДОВИ МОДЕЛЕЙ МАШИННОГО НАВЧАННЯУ роботі описані основні стратегії активного навчання, методи вибору об'єктів та основні прикладні задачі, в яких використання підходів активного навчання може суттєво зменшити вартість розмічування даних. Розглянуті варіанти використання методів активного навчання у поєднання з глибокими нейронними мережами.Ключові слова: активне навчання, human-in-the-loop, глибинне навчання.Nowadays, we have access to a huge amount of data that can be received from different sources: digitization of healthcare, the internet of things, social networks, online stores, and more others. And today the using of deep neural networks models for model creation has become incredibly popular. All these models require the availability of big data sets for training to be able to find hidden relationships between input data and output target variables, and this data for supervised learning tasks should be labeled. But the cost of data labeling in many cases can be quite high and may require the involvement of highly qualified experts. Therefore, there is a need to use active learning approaches, the main goal of which is to reduce the cost of data labeling due to the directed selection of objects of an unlabeled data set, which allows to increase the accuracy of machine learning models, while reducing the cost of data labeling. The goal of this article is the survey of such existing approaches and applied areas for using them. Three main scenarios of requests for unlabeled objects were observed in this work: pool-based sampling, streambased selective sampling, membership query synthesis. Among the methods of object selection, most popular methods were chosen, they are uncertainty sampling, Query-By-Committee, expected model change, variance reduction, estimated error reduction. Also, in this article different areas of using active learning were observed, such as medical image analysis, tasks of ranking search results, as well as the approaches to improving active learning methods.

show abstract

“…ECG CAD systems can serve as a valuable tool for medical professionals, facilitating objective diagnosis [3]. The association between different ECG records can be established through supervised [7], semisupervised [8], or unsupervised [9] ML approaches. Supervised learning entails training a model on a labeled dataset where ground-truth labels are known for each record.…”

Section: Introductionmentioning

confidence: 99%

“…Lower-level layers typically extract fundamental features like edges and color, while higher layers progressively abstract these features into semantically meaningful representations of the input [21]. This characteristic has spurred an active research field exploring the transferability of knowledge gained from pretrained models to the domain of ECG arrhythmia detection [8,22]. Alternatively, pretrained CNN models can be utilized as unsupervised feature extractors, bypassing the need for fine-tuning [23].…”

Section: Introductionmentioning

confidence: 99%

“…In contrast to CML-based approaches, the alternative pipeline leverages DL, which has emerged as the prevailing paradigm for tackling machine vision tasks [54]. CNNs constitute a prevalent DL architecture frequently employed for ECG data analysis, e.g., in [3][4][5]8,[16][17][18][19][20]55,56]. Unlike CML methods that rely on hand-engineered features, CNNs directly process raw ECG data as input.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Unraveling Arrhythmias with Graph-Based Analysis: A Survey of the MIT-BIH Database

Alinsaif

2024

Computation

View full text Add to dashboard Cite

Cardiac arrhythmias, characterized by deviations from the normal rhythmic contractions of the heart, pose a formidable diagnostic challenge. Early and accurate detection remains an integral component of effective diagnosis, informing critical decisions made by cardiologists. This review paper surveys diverse computational intelligence methodologies employed for arrhythmia analysis within the context of the widely utilized MIT-BIH dataset. The paucity of adequately annotated medical datasets significantly impedes advancements in various healthcare domains. Publicly accessible resources such as the MIT-BIH Arrhythmia Database serve as invaluable tools for evaluating and refining computer-assisted diagnosis (CAD) techniques specifically targeted toward arrhythmia detection. However, even this established dataset grapples with the challenge of class imbalance, further complicating its effective analysis. This review explores the current research landscape surrounding the application of graph-based approaches for both anomaly detection and classification within the MIT-BIH database. By analyzing diverse methodologies and their respective accuracies, this investigation aims to empower researchers and practitioners in the field of ECG signal analysis. The ultimate objective is to refine and optimize CAD algorithms, ultimately culminating in improved patient care outcomes.

show abstract

Semi-supervised active transfer learning for fetal ECG arrhythmia detection

Cited by 18 publications

References 22 publications

Machine Learning-Based Virtual Screening and Molecular Simulation Approaches Identified Novel Potential Inhibitors for Cancer Therapy

Machine Learning-Based Virtual Screening and Molecular Simulation Approaches Identified Novel Potential Inhibitors for Cancer Therapy

Використання Підходів Активного Навчання Під Час Побудови Моделей Машинного Навчання

Unraveling Arrhythmias with Graph-Based Analysis: A Survey of the MIT-BIH Database

Contact Info

Product

Resources

About