Figure - available from: Cluster Computing
This content is subject to copyright. Terms and conditions apply.
b Post-processing approach, this approach uses machine learning or other methods to process the result of basic CCA recognition method
Source publication
Canonical Correlation Analysis (CCA) is a popular way to analyze the underlying frequency components of an electroencephalogram (EEG) signal that contains Steady-State Visual Evoked Potentials (SSVEP). But solely itself may not be significant to detect the SSVEP frequency correctly. To improve its accuracy, several methods for processing the signal...
Citations
... BCIs of the first kind are reactive and active. An active BCI uses patterns in brain activity that the user directly and consciously controls to operate a device, regardless of what happens outside the device [5]. Reactive BCI harvests brain activity responses to environmental stimuli, which are then indirectly controlled by the user to operate a device. ...
The combination of brain cells and artificial intelligence (AI) is a paradigm shift in the healthcare industry that provides previously unheard-of chances for creativity and change in a variety of fields. This work is an attempt to offer a thorough examination of the confluence of AI and brain cells in healthcare, clarifying important ideas, methods, and applications that will influence medical practice and research going forward. Theis article provides an overview of AI in healthcare and looks at the wide variety of AI methods and algorithms advancing personalized medicine, therapy optimization, and disease diagnostics. It also touches upon how AI and brain cells interact, and how brain–computer interfaces (BCIs) can transform neuroscience research and human–machine interaction. It also highlights the revolutionary influence of brain cells and AI on healthcare delivery and patient care by outlining the application domains of the BCI across research fields and talking about the integration of reinforcement learning with the BCIs. It also showcases the practical applications of brain cells and AI in healthcare, ranging from prognostication and diagnostics to prosthetics and rehabilitation. This work suggests new trends and research and development opportunities in the field of brain cells and AI integration, as well as future directions in this field.
... Thoughtful use of multi-channel information can enhance the accuracy and robustness of SSVEP recognition. Therefore, most scholars conduct SSVEP researches based on multi-channel EEG signals [23][24][25]. Currently, there are two main categories for SSVEP recognition: traditional methods based on pre-extracted features and modern techniques for deep learning [26,27]. Deep learning-based methods have been becoming increasingly popular in recent years due to their high recognition accuracy, but they require high equipment requirements, complex model design, and a large amount of data to train the network. ...
Because of significant individual differences of brain signals, ensuring accuracy and information transfer rate (ITR) remains a challenge for SSVEP recognition. To address this issue, this study combines innovatively sliding time window method based on covariance analysis with the Task-related Component Analysis (TRCA). Moreover, the adaptive analysis technique is used to match different individuals by adjusting the initial length of time window and changing correlation analysis in TRCA. The results reveal that the smaller the initial time window length, the better the SSVEP recognition performance. With the fixed time window, TRCA using canonical correlation analysis (CCA) exhibits lower accuracy and smaller ITR compared to TRCA using Pearson correlation coefficient (PCC). However, under the sliding time window, TRCA with CCA finishes identification with 97% accuracy comparable to TRCA with PCC, and gets ITR up 20.7%. Furthermore, its accuracy and ITR are better than CCA, FBCCA, ABFCCA, and FBCCA-DW. The above results indicate that, during SSVEP identification, the sliding time window method based on covariance analysis is better paired with CCA than PCC, and the fixed time window method is better paired with PCC than CCA. Code is available.
... In this regard, there are some healthcare-related cases in which these algorithms can be used more specifically; for instance, Bayesian inference with canonical correlation analysis (CCA) can improve the SSVEP detection accuracy in EEG signals by means of finding steady-state visual evoked potentials (SSVEP) using Bayesian Inference. These applications have potential to improve recognition rate significantly and also it can optimize healthcare delivery through non-invasive brain-computer interfaces [13]. ...
The study has two objectives: foremost, it is imperative to establish likely roles for social robots in facilitating collaboration and interaction among stakeholders such as the senior citizens, caregivers, health care professionals and traditional Chinese medicine (TCM) practitioners. Secondly, mathematical fusion algorithms discussion capable of integrating data from multiple sensors is meant to enhance perception capacity of social robots in select care settings while promoting value co-creation in existing networks for elderly care. In that regard, therefore, it is about exploring how quality care services can be improved through the use of social robots coupled with mathematical fusion algorithms; thus looking at effective communication processes as well as creating a collaborative environment among stakeholders involved. Meanwhile, applying social robotics together with multisensor integration within elder care networks located in Wenzhou city may have feasibility, effectiveness and potential benefits hence leading improved overall wellbeing satisfaction levels among its aging residents.
