Publications by authors named "O Yamashita"
Sparse Bayesian learning has promoted many effective frameworks of brain activity decoding for the brain-computer interface, including the direct reconstruction of muscle activity using brain recordings. However, existing sparse Bayesian learning algorithms mainly use Gaussian distribution as error assumption in the reconstruction task, which is not necessarily the truth in the real-world application. On the other hand, brain recording is known to be highly noisy and contains many non-Gaussian noises, which could lead to large performance degradation for sparse Bayesian learning algorithms.
View Article and Find Full Text PDFArticle Synopsis
- Autism spectrum disorder (ASD) is a complex lifelong condition, and this study aimed to create a classifier using resting-state fMRI from a large group of 730 Japanese adults to identify its neural and biological features.
- The developed classifier showed effectiveness in differentiating individuals with ASD from neurotypical controls across various countries, including the US and Belgium, and it also applied to children and adolescents.
- Importantly, the study found that the classifier identified crucial functional connections related to social interaction difficulties and neurotransmitter activity, and it linked ASD with similar neurobiological factors seen in ADHD and schizophrenia, enhancing understanding of related mental health disorders.
Article Synopsis
- Neuroimaging databases for neuro-psychiatric disorders provide valuable data for researchers to explore diseases, develop machine learning models, and redefine understanding of these conditions.* ! -
- A review identified 42 global MRI datasets totaling 23,293 samples from patients with various disorders, including mood, developmental, schizophrenia, Parkinson's, and dementia.* ! -
- Improved governance and addressing technical issues of these databases are essential for sharing data across borders, aiding in understanding, diagnosing, and creating early interventions for neuro-psychiatric disorders.* !
Dynamic mode (DM) decomposition decomposes spatiotemporal signals into basic oscillatory components (DMs). DMs can improve the accuracy of neural decoding when used with the nonlinear Grassmann kernel, compared to conventional power features. However, such kernel-based machine learning algorithms have three limitations: large computational time preventing real-time application, incompatibility with non-kernel algorithms, and low interpretability.
View Article and Find Full Text PDFDeep learning is a powerful tool for neural decoding, broadly applied to systems neuroscience and clinical studies. Interpretable and transparent models that can explain neural decoding for intended behaviors are crucial to identifying essential features of deep learning decoders in brain activity. In this study, we examine the performance of deep learning to classify mouse behavioral states from mesoscopic cortex-wide calcium imaging data.
View Article and Find Full Text PDF