Original title: Functional near-infrared spectroscopy pre-ictal oscillations recorded simultaneously with stereo-EEG
Netaniel Rein, Revital Shechter, Yael Avni, Evgeny Tsizin +13 · September 2026
Doctors want to know if brain blood flow changes before an epileptic seizure begins. Functional near-infrared spectroscopy, or fNIRS, uses light shone through the scalp to track blood oxygen levels. Researchers matched these optical signals with deep brain electrodes, called SEEG, which pinpoint the exact second an electrical seizure starts. They examined ultra-slow waves in blood flow that cycle once every several minutes. The team tested whether these slow waves appeared before actual seizures more often than during random control periods. They found that these ultra-slow blood flow shifts did rise before electrical seizures started, showing up hundreds of seconds ahead of time. These changes clustered near where the seizures began and did not match shallow signals from the scalp. This reveals that blood flow shifts can start well ahead of the electrical seizure, pointing to slow vascular events linked to how seizures develop.
Imagine you are listening to a quiet room, waiting for a friend to speak. You hear a faint rustle that is too low to be a voice, but it happens regularly before each outburst. The researchers used light to listen to the brain's blood flow, and they found a similar ultra-low hum that begins minutes before a seizure. In this paper, the ultra-slow fluctuations (the faint rustle) precede the seizure (the friend speaking) and are not random noise.
For people with epilepsy, seizures can strike without warning. This study shows that a non-invasive light-based technique can detect brain activity changes up to ten minutes before seizure onset. If confirmed, this could lead to wearable devices that predict seizures and give patients time to prepare.
The study involved only five patients and sixteen seizures, so results may not generalize. The ultra-slow frequency band (0.002-0.004 Hz) was chosen from exploratory inspection and requires independent validation. The physiological origin and clinical significance of these oscillations remain unknown. Not all seizures showed the effect, and the statistical significance was limited to one channel after correction.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Functional near-infrared spectroscopy pre-ictal oscillations recorded simultaneously with stereo-EEG — Netaniel Rein, Revital Shechter, Yael Avni et al. — Neuroimage. Reports — doi:10.1016/j.ynirp.2026.100401
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper.