FaceID共同創業者、脳波からうつ病を診断するフロンティアモデルを訓練

hemispheric has secured $52 million in funding to advance a frontier ai model trained on a dataset of 250,000 hours of brain electrical activity collected from 100,000 volunteers, enabling the diagnosis of conditions such as alzheimer's, schizophrenia, and depression through a standard eeg headset. founded by gidi littwin, a co-inventor of apple's faceid and vision pro technology, the company applies deep learning techniques originally used for facial recognition to decode brain signals, aiming to bypass the invasive procedures and subjective questionnaires currently required in psychiatry. with clinical validation underway, hemispheric intends to file for fda approval for its ptsd diagnostic tool in early 2027, positioning itself to offer a low-cost, scalable solution for mental health clinics that could democratize access to neurological diagnostics. this development signals a growing trend of ai convergence with healthcare, where specialized 'frontier' models are beginning to challenge general-purpose models from giants like openai and anthropic in specific medical applications.