Love the progress in brain imaging. Similar to fMRI, this seems to be focused on identifying blood flow to different areas which seems useful for a lot of diagnostic/medical purposes. I’m personally doubtful that the use cases at the end of the article like brain-computer interfaces are well-suited to the technology though, since it can’t measure neural activity as directly as electrodes.

One thing it doesn’t discuss is the latency of the observed signals relative to brain activity — in MRI there is a 5-6 second delay due to the time it takes for brain activity to translate into blood flow and oxygen signals, whereas electrical activity is in real time.