Media & Culture

Hemispheric's AI trained on 100,000 brains to diagnose cognitive disorders

FaceID co-inventor's startup raises $52M to decode brain activity with AI

Deep Dive

Gidi Littwin, co-inventor of Apple's FaceID and Vision Pro, has spent six years building Hemispheric, a startup that uses frontier AI to decode electrical brain activity and diagnose cognitive disorders. The company has raised $52 million in early-stage funding from investors including Howard Morgan. Littwin and cofounder Hagai Lalazar collected a quarter-million hours of brain data from 100,000 paid volunteers across Asia, Tel Aviv, and Boston. Subjects played game-like activities that activated different brain regions, and the resulting data trained a deep-learning model that infers brain function from EEG signals, similar to how LLMs deduce meaning from text.

The model has been tested on individuals with PTSD, schizophrenia, and depression, making accurate deductions about brain health. Hemispheric plans to submit its first product—a PTSD diagnostic tool—to the FDA for approval early next year, with public availability by 2027. Patients will wear a lightweight EEG headset for 15 minutes while interacting with a tablet app, and the AI will help clinicians diagnose, predict treatment response, and monitor progress. The company is also developing its own brain scanners optimized for machine learning, arguing that traditional EEGs were never built for deep learning. Hemispheric aims to make its device as common as a blood test in mental health clinics.

Key Points
  • Raised $52M from investors including early Uber-backer Howard Morgan
  • Trained on 250,000 hours of EEG data from 100,000 paid volunteers
  • First product targets PTSD; FDA submission early 2025, public rollout by 2027

Why It Matters

AI-powered brain diagnostics could replace subjective questionnaires, enabling early detection of Alzheimer's, depression, and PTSD.

📬 Get the top 10 AI stories daily