# EPFL's LogSync implant runs on-chip machine learning at 200 microwatts to trigger brain stimulation only when it detects a target state

The 4-by-2-millimeter chip reads synchrony between brain regions to decide when to stimulate, but its human results come from recorded datasets, not implants.

- Published: 2026-10-29T06:14:22.974Z
- Canonical: https://polylog.news/ai/2026-10-29/epfl-s-logsync-implant-runs-on-chip-machine-learning-at-200
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news), [News-Medical](https://news-medical.net/news/20261023/Low-power-implantable-neural-chip-tracks-communication-across-brain-regions.aspx), [Neuroscience News](https://neurosciencenews.com/dbs-depression-neurotech-32338/)

Engineers at the Swiss Federal Institute of Technology in Lausanne (EPFL) built LogSync, an implantable system-on-chip that tracks communication between brain regions, the Russian-language channel AI ML Big Data reported. Mahsa Shoaran, hea…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-10-29/epfl-s-logsync-implant-runs-on-chip-machine-learning-at-200 (subscription information: https://polylog.news/pricing).