Dataset Information

A filter approach for feature selection in classification: application to automatic atrial fibrillation detection in electrocardiogram recordings.

ABSTRACT:

Background

In high-dimensional data analysis, the complexity of predictive models can be reduced by selecting the most relevant features, which is crucial to reduce data noise and increase model accuracy and interpretability. Thus, in the field of clinical decision making, only the most relevant features from a set of medical descriptors should be considered when determining whether a patient is healthy or not. This statistical approach known as feature selection can be performed through regression or classification, in a supervised or unsupervised manner. Several feature selection approaches using different mathematical concepts have been described in the literature. In the field of classification, a new approach has recently been proposed that uses the [Formula: see text]-metric, an index measuring separability between different classes in heart rhythm characterization. The present study proposes a filter approach for feature selection in classification using this [Formula: see text]-metric, and evaluates its application to automatic atrial fibrillation detection.

Methods

The stability and prediction performance of the [Formula: see text]-metric feature selection approach was evaluated using the support vector machine model on two heart rhythm datasets, one extracted from the PhysioNet database and the other from the database of Marseille University Hospital Center, France (Timone Hospital). Both datasets contained electrocardiogram recordings grouped into two classes: normal sinus rhythm and atrial fibrillation. The performance of this feature selection approach was compared to that of three other approaches, with the first two based on the Random Forest technique and the other on receiver operating characteristic curve analysis.

Results

The [Formula: see text]-metric approach showed satisfactory results, especially for models with a smaller number of features. For the training dataset, all prediction indicators were higher for our approach (accuracy greater than 99% for models with 5 to 17 features), as was stability (greater than 0.925 regardless of the number of features included in the model). For the validation dataset, the features selected with the [Formula: see text]-metric approach differed from those selected with the other approaches; sensitivity was higher for our approach, but other indicators were similar.

Conclusion

This filter approach for feature selection in classification opens up new methodological avenues for atrial fibrillation detection using short electrocardiogram recordings.

SUBMITTER: Michel P

PROVIDER: S-EPMC8094578 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:To develop electrocardiogram (ECG) tools to quantify the number of sources for atrial fibrillation (AF), i.e. spatially stable rotors and focal impulses, and whether they lie in right or left atrium. Intracardiac mapping has recently shown that paroxysmal and persistent AF is sustained by rotors or focal sources that are stable in location and thus targets for limited ablation [focal impulse and rotor modulation (FIRM)] to eliminate AF. Importantly, the numbers and locations of concurrent sources determine both the complexity of AF and the approach for ablation.In 36 AF patients (n = 29 persistent, 63 ± 9 years) in the CONventional ablation with or without Focal Impulse and Rotor Modulation (CONFIRM) trial, we developed phase lock (PL) to quantify spatial repeatability of ECG 'F-waves' between leads over time. Phase lock spectrally quantifies the angle ? between F-wave voltages in planes formed by ECG leads I, aVF, and V1 at successive points in time. We compared PL with ECG spectral dominant frequency (DF) and organizational index (OI) to characterize stable rotors and focal sources validated by intracardiac FIRM mapping. Focal impulse and rotor modulation ablation alone at ?3 sources acutely terminated and rendered AF non-inducible or substantially slowed AF in 31 of 36 patients. Receiver operating characteristics of PL for this endpoint had area under the curve (AUC) = 0.72, and the optimum cut-point (PL = 0.09) had 74% sensitivity, 92% positive predictive value (PPV). Receiver operating characteristics areas for OI and DF were 0.50 and 0.58, respectively. Left (n = 28) or right (n = 3) atrial sources were localized by PL with AUC = 0.85, sensitivity 100%, PPV 30%, and negative predictive value 100%. Spectral DF provided AUC = 0.79. Notably, PL did not comigrate with diagnosis of paroxysmal or persistent AF (P = NS), unlike ECG DF.The novel metric of ECG PL identifies patients with fewer (?3) or greater numbers of stable rotors/focal sources for AF, validated by intracardiac FIRM mapping, and localized them to right or left atria. These data open the possibility of using 12-lead ECG analyses to classify AF mechanistically and plan procedures for right- or left-sided FIRM ablation.

Project description:BACKGROUND:MicroRNAs (miRNAs) are noncoding RNA molecules heavily involved in human tumors, in which few of them circulating the human body. Finding a tumor-associated signature of miRNA, that is, the minimum miRNA entities to be measured for discriminating both different types of cancer and normal tissues, is of utmost importance. Feature selection techniques applied in machine learning can help however they often provide naive or biased results. RESULTS:An ensemble feature selection strategy for miRNA signatures is proposed. miRNAs are chosen based on consensus on feature relevance from high-accuracy classifiers of different typologies. This methodology aims to identify signatures that are considerably more robust and reliable when used in clinically relevant prediction tasks. Using the proposed method, a 100-miRNA signature is identified in a dataset of 8023 samples, extracted from TCGA. When running eight-state-of-the-art classifiers along with the 100-miRNA signature against the original 1046 features, it could be detected that global accuracy differs only by 1.4%. Importantly, this 100-miRNA signature is sufficient to distinguish between tumor and normal tissues. The approach is then compared against other feature selection methods, such as UFS, RFE, EN, LASSO, Genetic Algorithms, and EFS-CLA. The proposed approach provides better accuracy when tested on a 10-fold cross-validation with different classifiers and it is applied to several GEO datasets across different platforms with some classifiers showing more than 90% classification accuracy, which proves its cross-platform applicability. CONCLUSIONS:The 100-miRNA signature is sufficiently stable to provide almost the same classification accuracy as the complete TCGA dataset, and it is further validated on several GEO datasets, across different types of cancer and platforms. Furthermore, a bibliographic analysis confirms that 77 out of the 100 miRNAs in the signature appear in lists of circulating miRNAs used in cancer studies, in stem-loop or mature-sequence form. The remaining 23 miRNAs offer potentially promising avenues for future research.

Project description:BACKGROUND:Increasing prevalence of atrial fibrillation has a significant impact on health, society, and healthcare resource utilization, due to increased morbidity, mortality, risk of stroke, and reduction in quality of life. Early diagnosis allows for treatment initiation, a reduction in complications and associated costs, and so innovation to improve screening and enable easy access are needed Developments in digital technology have significantly contributed to the availability of screening tools. The single-lead electrocardiogram AliveCor (Mountainview, CA) device offers the opportunity to provide heart rhythm screening and has been used extensively in clinical practice and research studies. METHODS:This review investigates the feasibility, validity, and utility of the AliveCor device as a tool for atrial fibrillation detection in clinical practice and in wider research. Databases searched included PUBMED, CINAHL, MEDLINE, and World of Science, plus grey literature search. Search terms related to atrial fibrillation, screening, and AliveCor with adults >18 years. Feasibility metrics were applied including process, resource, management, and scientific outcomes. Studies not written in the English language were excluded. Validity of AliveCor was explored by extracting sensitivity and specificity data from eligible studies and overall effectiveness analyzed by incorporating the above, with wider issues surrounding screening approaches, cost effectiveness and appropriateness of AliveCor as a screening tool. RESULTS:The AliveCor device screening was reviewed in 11 studies matching inclusion criteria. Atrial fibrillation detection rates ranged from 0.8% to 36% and this largely correlated to the study population, where wider age inclusion and mass/population screening represented lower atrial fibrillation detection. Recruitment from higher-risk groups (older age, targeted localities, chronic disease) identified higher numbers with atrial fibrillation. Feasibility metrics demonstrated AliveCor as an effective tool of choice in terms of process, resources, and management. Duration of screening time had an impact on rates of atrial fibrillation detection. There was however significant heterogeneity between studies reviewed. CONCLUSION:The AliveCor device offers a convenient, valid, and feasible means of monitoring for atrial fibrillation. Further analysis of electrocardiograms produced by AliveCor may be necessary in some circumstances. The AliveCor electrocardiogram device can be successfully implemented into both opportunistic and systematic screening strategies for atrial fibrillation.

Project description:BackgroundAtrial fibrillation (AFib) is the most common cardiac arrhythmia associated with stroke, blood clots, heart failure, coronary artery disease, and/or death. Multiple methods have been proposed for AFib detection, with varying performances, but no single approach appears to be optimal. We hypothesized that each state-of-the-art algorithm is appropriate for different subsets of patients and provides some independent information. Therefore, a set of suitably chosen algorithms, combined in a weighted voting framework, will provide a superior performance to any single algorithm.MethodsWe investigate and modify 38 state-of-the-art AFib classification algorithms for a single-lead ambulatory electrocardiogram (ECG) monitoring device. All algorithms are ranked using a random forest classifier and an expert-labeled training dataset of 2,532 recordings. The seven top-ranked algorithms are combined by using an optimized weighting approach.ResultsThe proposed fusion algorithm, when validated on a separate test dataset consisting of 4,644 recordings, resulted in an area under the receiver operating characteristic (ROC) curve of 0.99. The sensitivity, specificity, positive-predictive-value (PPV), negative-predictive-value (NPV), and F1-score of the proposed algorithm were 0.93, 0.97, 0.87, 0.99, and 0.90, respectively, which were all superior to any single algorithm or any previously published.ConclusionThis study demonstrates how a set of well-chosen independent algorithms and a voting mechanism to fuse the outputs of the algorithms, outperforms any single state-of-the-art algorithm for AFib detection. The proposed framework is a case study for the general notion of crowdsourcing between open-source algorithms in healthcare applications. The extension of this framework to similar applications may significantly save time, effort, and resources, by combining readily existing algorithms. It is also a step toward the democratization of artificial intelligence and its application in healthcare.

Project description:ObjectivesWidespread implementation of electronic databases has improved the accessibility of plaintext clinical information for supplementary use. Numerous machine learning techniques, such as supervised machine learning approaches or ontology-based approaches, have been employed to obtain useful information from plaintext clinical data. This study proposes an automatic multi-class classification system to predict accident-related causes of death from plaintext autopsy reports through expert-driven feature selection with supervised automatic text classification decision models.MethodsAccident-related autopsy reports were obtained from one of the largest hospital in Kuala Lumpur. These reports belong to nine different accident-related causes of death. Master feature vector was prepared by extracting features from the collected autopsy reports by using unigram with lexical categorization. This master feature vector was used to detect cause of death [according to internal classification of disease version 10 (ICD-10) classification system] through five automated feature selection schemes, proposed expert-driven approach, five subset sizes of features, and five machine learning classifiers. Model performance was evaluated using precisionM, recallM, F-measureM, accuracy, and area under ROC curve. Four baselines were used to compare the results with the proposed system.ResultsRandom forest and J48 decision models parameterized using expert-driven feature selection yielded the highest evaluation measure approaching (85% to 90%) for most metrics by using a feature subset size of 30. The proposed system also showed approximately 14% to 16% improvement in the overall accuracy compared with the existing techniques and four baselines.ConclusionThe proposed system is feasible and practical to use for automatic classification of ICD-10-related cause of death from autopsy reports. The proposed system assists pathologists to accurately and rapidly determine underlying cause of death based on autopsy findings. Furthermore, the proposed expert-driven feature selection approach and the findings are generally applicable to other kinds of plaintext clinical reports.