We revisit a previous study on inter-session variability (McGonigle et al. [2000]: Neuroimage 11:708-734), showing that contrary to one popular interpretation of the original article, inter-session variability is not necessarily high. We also highlight how evaluating variability based on thresholded single-session images alone can be misleading. Finally, we show that the use of different first-level preprocessing, time-series statistics, and registration analysis methodologies can give significantly different inter-session analysis results.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6871748 | PMC |
| http://dx.doi.org/10.1002/hbm.20080 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Systematic review of experimental paradigms and deep neural networks for electroencephalography-based cognitive workload detection.
Prog Biomed Eng (Bristol)
October 2024
Neural Engineering Lab, Department of Biosciences and Bioengineering, Indian Institute of Technology Guwahati, Assam 781039, India.
This article summarizes a systematic literature review of deep neural network-based cognitive workload (CWL) estimation from electroencephalographic (EEG) signals. The focus of this article can be delineated into two main elements: first is the identification of experimental paradigms prevalently employed for CWL induction, and second, is an inquiry about the data structure and input formulations commonly utilized in deep neural networks (DNN)-based CWL detection. The survey revealed several experimental paradigms that can reliably induce either graded levels of CWL or a desired cognitive state due to sustained induction of CWL.
View Article and Find Full Text PDFFeasibility, reliability and validity of self-measurement of knee range-of-motion using an accelerometer-based smartphone application by patients with total knee arthroplasty.
PLoS One
October 2024
Department of Physiotherapy, Singapore General Hospital, Singapore, Singapore.
Article Synopsis
- After knee surgery, many people struggle to move their knees properly, which can make them unhappy and limit their activities.
- This study looked at whether patients could accurately measure their knee movement using a special app on their smartphones instead of needing a healthcare professional.
- The results showed that almost all patients could use the app easily, and it gave very reliable measurements of knee movement.
Improving inter-session performance via relevant session-transfer for multi-session motor imagery classification.
Heliyon
September 2024
Department of Biomedical Engineering, Korea University College of Medicine, Seoul, 02841, Republic of Korea.
Article Synopsis
- - This study addresses challenges in motor imagery (MI)-based brain-computer interfaces (BCIs) caused by non-stationary EEG signals that affect performance across different sessions.
- - The researchers introduced a novel technique called the relevant session-transfer (RST) method, which improves MI classification by transferring relevant EEG data from previous sessions to the current one.
- - Experiments showed that the RST method improved classification performance by 2.29% on a public dataset and 6.37% on a customized dataset, outperforming other methods and confirming its effectiveness for multi-session applications in MI-BCIs.
Repeatability of gait of children with spastic cerebral palsy in different walking conditions.
J Biomech
November 2024
KU Leuven, Department of Rehabilitation Sciences, Leuven, Belgium; Clinical Motion Analysis Laboratory, University Hospitals Leuven, Leuven, Belgium.
Three-dimensional gait analysis is the 'gold standard' for measurement and description of gait. Gait variability can arise from intrinsic and extrinsic factors and may vary between walking conditions. This study aimed to define the inter-trial and inter-session repeatability in gait analysis data of children with cerebral palsy (CP) who were walking in four conditions, namely barefoot or with ankle-foot orthosis (AFO), and overground or treadmill.
View Article and Find Full Text PDFAn Ultra-Low Power Wearable BMI System with Continual Learning Capabilities.
IEEE Trans Biomed Circuits Syst
September 2024
Driven by the progress in efficient embedded processing, there is an accelerating trend toward running machine learning models directly on wearable Brain-Machine Interfaces (BMIs) to improve portability and privacy and maximize battery life. However, achieving low latency and high classification performance remains challenging due to the inherent variability of electroencephalographic (EEG) signals across sessions and the limited onboard resources. This work proposes a comprehensive BMI workflow based on a CNN-based Continual Learning (CL) framework, allowing the system to adapt to inter-session changes.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!