Key result
The F-score_SVM model combined with a novel spatial denoising algorithm achieved 96.11% sensitivity and 96.05% specificity for P300-based lie detection, outperforming models without spatial denoising.
Why the study?
Does a novel spatial denoising algorithm combined with F-score and SVM improve classification accuracy of P300 potentials for lie detection in healthy subjects?
Population
34 healthy subjects recruited from a university.
Comparison
Spatial denoising algorithm based on independent… vs Results obtained without using SDA and results…
Design
Other, Thirty-four subjects were divided randomly into guilty and innocent…
Authors
Loading...
May enhance P300 lie detection accuracy; leaves open validation in larger prospective studies.
Does a novel spatial denoising algorithm combined with F-score and SVM improve classification accuracy of P300 potentials for lie detection in healthy subjects?
A novel spatial denoising algorithm combined with F-score feature selection and SVM classification significantly enhances P300 signal-to-noise ratio and improves lie detection accuracy using a small number of stimuli.
Gao et al. (2014) studied Lie Detection (n=34). F-score_SVM with Spatial Denoising Algorithm (SDA) vs. Classification without SDA and other models (FDA, BPNN) was evaluated on Classification accuracy (sensitivity and specificity). The F-score_SVM model combined with a novel spatial denoising algorithm achieved 96.11% sensitivity and 96.05% specificity for P300-based lie detection, outperforming models without spatial denoising.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: