Key result
The proposed hybrid bio-inspired incremental learning-based model improved the accuracy of meditation detection by 5.9% and effect analysis accuracy by 8.3% compared to existing methods.
Why the study?
Existing models for analyzing meditation effects are complex, have lower clinical efficiency, and rely on single analysis types, restricting their applicability.
A novel hybrid bio-inspired incremental learning model using EEG signals improves the accuracy of detecting meditation and analyzing its effects on the body compared to existing methods.
May aid EEG-based meditation research; leaves open cardiovascular applications pending clinical validation.
Meditation has proven to be a highly effective technique for improving the health conditions of different body parts. Pranayama assists in improving lung & brain function, while Transcendental Meditation assists in reducing stress and improving heart functions. Thus, effective analysis of meditation & its effects on human body can assist in intelligent adaptation of these techniques for different disorders. A wide variety of scanning, pre-processing & classification models are proposed by researchers to perform these tasks, and each of these models vary in terms of their application & function characteristics. But these models are either highly complex, or have lower efficiency when used under clinical conditions. Moreover, most of these models use a single type of analysis (either frequency, or entropy, or their counterparts), which restricts their applicability to limited use cases. To overcome these issues, this text proposes design of a novel hybrid bio inspired incremental learning-based model for analyzing effects of meditation on different body-parts. The proposed model uses a collected of pre-stored EEG signals of persons who have meditated & not meditated, along with their temporal health-reports. These signals are initially represented in multiple domains that assists in estimation of frequency, entropy, Gabor, wavelet, and convolutional feature sets. These feature sets are later selected by an Elephant Herding optimizer (EHO) which assists in identification of highly variant features. These features are used to train a customized ID CNN for identifying effects of meditation on different body parts, on temporal basis. Due to these operations, the proposed model is able to improve accuracy of meditation detection by 5.9%, and improve the effect analysis accuracy by 8.3% when compared with existing methods under similar use cases.
No takes yet. Share an insight, caveat, or question.
Tayade et al. (2023) studied Meditation effects. Hybrid bio inspired incremental learning-based model (BILMB) vs. Existing methods was evaluated on Accuracy of meditation detection and effect analysis (5.9% improvement in detection, 8.3% in effect analysis). The proposed hybrid bio-inspired incremental learning-based model improved the accuracy of meditation detection by 5.9% and effect analysis accuracy by 8.3% compared to existing methods.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: