A hierarchical meta-model for multi-class mental task based brain-computer interfaces

Gupta, Akanksha; Agrawal, R. K.; Kirar, Jyoti Singh; Kaur, Baljeet; Ding, Weiping; Lin, Chin‐Teng; Andreu-Pérez, Javier; Prasad, Mukesh

doi:10.1016/j.neucom.2018.07.094

Cited by 9 publications

(7 citation statements)

References 36 publications

(58 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed CNN framework can also be used in material informatics [77, 78, 79, 80, 81, 82]. Nevertheless, the proposed MSFFCNN model can be employed as a more reliable and robust MI-based real-time BCI applications such as robotic control [9, 10, 11], rehabilitation of neuromotor disorders [8], text entry speech communication [12, 13] etc.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

A multi-scale fusion CNN model based on adaptive transfer learning for multi-class MI-classification in BCI system

Roy

2022

Preprint

View full text Add to dashboard Cite

Deep learning-based brain-computer interface (BCI) in motor imagery (MI) has emerged as a powerful method for establishing direct communication between the brain and external electronic devices. However, due to inter-subject variability, inherent complex properties, and low signal-to-noise ratio (SNR) in electroencephalogram (EEG) signal are major challenges that significantly hinder the accuracy of the MI classifier. To overcome this, the present work proposes an efficient transfer learning-based multi-scale feature fused CNN (MSFFCNN) which can capture the distinguishable features of various non-overlapping canonical frequency bands of EEG signals from different convolutional scales for multi-class MI classification. In order to account for inter-subject variability from different subjects, the current work presents 4 different model variants including subject-independent and subject-adaptive classification models considering different adaptation configurations to exploit the full learning capacity of the classifier. Each adaptation configuration has been fine-tuned in an extensively trained pre-trained model and the performance of the classifier has been studied for vast range of learning rates and degrees of adaptation which illustrates the advantages of using an adaptive transfer learning-based model. The model achieves an average classification accuracy of 94.06 % ±2.29 % and kappa value of 0.88 outperforming several baseline and current state-of-the-art EEG-based MI classification models with fewer training samples. The present research provides an effective and efficient transfer learning-based end-to-end MI classification framework for designing a high-performance robust MI-BCI system.

show abstract

Section: Discussionmentioning

confidence: 99%

“…commands to control the external electronic devices [1,2,3,4,5,6,7]. The BCI allows rehabilitation of neuromotor disorders [8], robotic control [9,10,11], speech communication [12,13], etc. In BCI paradigms, MI classification is the most critical part in which brain signals can be translated into control signals [14,15].…”

mentioning

confidence: 99%

A multi-scale fusion CNN model based on adaptive transfer learning for multi-class MI-classification in BCI system

Roy

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…A binary class mental task classification is extended to a multi class mental problem by Akshansh Gupta et al [5] wherein a sets of relevant and non-redundant features are selected using a linear regression and a multivariate feature selection. At the last an optimal decision tree based support vector machine classifier was used for multimental task classification.…”

Section: Brain Computer Interfacementioning

confidence: 99%

In situ developmental control of biomedical applications using human brain

S¹

2020

IJATCSE

View full text Add to dashboard Cite

This article describes the development of BCI (Brain computer Interface) that allows individuals to record their brain activity and then utilize it for useful application. A virtual reality system is connected with EEG device for smart home application. It paves way for usual way of control and communication. Its an artificial system that sidesteps the neuromuscular output terminals which are efficient pathways of our body. Different brain states resulting due to neural interactions lead to the development of patterns which are characterized by different amplitudes and frequencies. A minute electrical discharge is observed for every interaction between neurons received by brain which is separated into packets which are transmitted through a remoteway of communication (Bluetooth). The wave measuring unit will receive the immediate first information of the brain wave directly through Arduino Uno Controller. Next the instruction is sent to the home section. Taking a notice of the brain action, certain thoughts and eye squinting pattern, relationship were established between exchanging and regulation of certain home appliances.

show abstract

“…To increase the visibility/differentiating impact of the current issue, we summarize the salient features of the relevant methods in Table 1. Besides the work mentioned above, researchers have employed fuzzy nonlinear modeling approaches in various applications [63][64][65][66][67].…”

Section: Introductionmentioning

confidence: 99%

Analytical Reduction Method for New Type-2 Fuzzy Chance-Constrained Portfolio Selection Model

Yang¹,

Mei²,

Qin³

et al. 2021

IJCIS

View full text Add to dashboard Cite

In the traditional portfolio selection problem, asset returns are modeled as fuzzy variables with fuzzy return. However, this approach is limited in its ability to capture uncertainty accurately and in analytical model solving. Here, we aim to develop a new fuzzy chance-constrained portfolio model with a type-2 fuzzy return variable using a credibility measure. In real practice, an effective portfolio model under a new, more complex environment is required to improve instinctive imprecision. Here, we propose a novel analytical reduction method to transform our proposed model into a linear programing model with linear constraints, and use a linear programing tool to obtain optimal portfolio strategies. We first reformulate the portfolio model with type-2 fuzzy returns using two types of chance criteria. Next, we provide a new analytical method to solve the proposed model. Then, we present a numerical example with 20 asset returns described by a triangular membership function and use comparison testing to illustrate the advantages of our proposed method. The numerical results show that the relationship between investor tolerance of portfolio risk and the values attained for the four objective functions is in line with our expectations regarding the risk-return trade-off, and the comparison test results indicate that our proposed reduction method performs better than three existing methods. Our method provides an effective practice model for reformulating type-2 fuzzy portfolio problems using an analytical reduction method. Although a large number of existing type-2 fuzzy portfolio problems cannot be solved by our analytical method, it represents a new tool to solve these kinds of problems.

show abstract

A hierarchical meta-model for multi-class mental task based brain-computer interfaces

Cited by 9 publications

References 36 publications

A multi-scale fusion CNN model based on adaptive transfer learning for multi-class MI-classification in BCI system

A multi-scale fusion CNN model based on adaptive transfer learning for multi-class MI-classification in BCI system

In situ developmental control of biomedical applications using human brain

Analytical Reduction Method for New Type-2 Fuzzy Chance-Constrained Portfolio Selection Model

Contact Info

Product

Resources

About