alternative to the research of brain dynamics

In order to identify differences across brain regions using functional magnetic resonance imaging (fMRI) data, Sip et al. (2023) devised a machine-learning approach. Since the method is data-driven and unsupervised, the model can determine regional properties without relying on previously established features of neuronal masses. The model's output was compared to acknowledged genetic and structural characteristics of the brain, which improved our comprehension of brain dynamics. The model can be expanded to include task-based scans or clinical data even though it was initially created on healthy people in resting condition. With less reliance on preconceived notions and more on unsupervised learning and raw observational data, this unique technique might provide a trustworthy alternative to the research of brain dynamics.

Reference: Sip, V., Hashemi, M., Dickscheid, T., Amunts, K., Petkoski, S., & Jirsa, V. (2023). Characterization of regional differences in resting-state fMRI with a data-driven network model of brain dynamics. Science Advances, 9(12), abq754. https://doi.org/10.1126/sciadv.abq754 

Related concepts: 

-The study of medical imaging data using machine learning

-Creation of unsupervised machine learning methods for complex systems research

-The use of artificial intelligence in neuroscience research to learn more about the brain and its processes.