Decoding neuropsychiatric states from multi-site brain network activity

    公开(公告)号:US12097029B1

    公开(公告)日:2024-09-24

    申请号:US16031925

    申请日:2018-07-10

    发明人: Maryam Shanechi

    摘要: A method for decoding mood or other neuropsychiatric states from large-scale brain activity includes receiving, by a processor, large-scale brain activity signals from an electrode assembly coupled to a subject. The method includes continuously and automatically decoding a neuropsychiatric state of the subject from the large-scale brain activity signals received from a predictive network subset of brain sites coupled to the electrode assembly, contemporaneously with the receiving. The method includes providing a signal indicative of the neuropsychiatric state. A method for adaptive tracking of large-scale brain network activity includes characterizing, by one or more computers, a time-variant linear state-space model predictive of a brain state, at least in part by updating estimates of time-varying covariance matrices at time steps. The method includes tracking large-scale brain network activity in a subject using an electrode array and continuously and automatically estimating the subject's brain state contemporaneously with the characterizing and tracking.

    Robust real-time EEG suppression detection device and method

    公开(公告)号:US12089942B1

    公开(公告)日:2024-09-17

    申请号:US18084615

    申请日:2022-12-20

    IPC分类号: A61B5/316 A61B5/00 A61B5/369

    摘要: The present invention relates to a physiological monitor and system, more particularly to an electroencephalogram (EEG) monitor and system, and a method of detecting the presence or occurrence of suppression in the EEG signal. Accurately detecting signal suppression in real-time provides the clinician with the ability to prevent possibly severe, long-term damage to patients as a result of excessive anesthetic or sedative. The present invention provides such a system and method for accurately and automatically detecting suppression in physiological, particularly EEG, signals in real-time and allowing for the administration of treatment or medication to reverse the effects of such situations, or minimize the harm caused. The present invention also allows for the use of closed-loop treatment or drug delivery systems to further automate the process and provide rapid treatment to a patient to reverse or minimize potential harm.

    Neural oscillation monitoring system

    公开(公告)号:US12076166B2

    公开(公告)日:2024-09-03

    申请号:US15257019

    申请日:2016-09-06

    申请人: Newton Howard

    发明人: Newton Howard

    摘要: Embodiments of the present invention may provide automated techniques for signal analysis that may continuously provide up-to-date results that link EEG and behaviors that are important for daily activities. Such techniques may provide automation, objectivity, real-time monitoring and portability. In an embodiment of the present invention, a computer-implemented method for monitoring neural activity may comprise receiving data representing at least one signal representing neural activity of a test subject, pre-processing the received data by performing at least one of band-pass filtering, artifact removal, identifying common spatial patterns, and temporally segmentation, processing the pre-processed data by performing at least one of time domain processing, frequency domain processing, and time-frequency domain processing, generating a machine learning model using the processed data as a training dataset, and outputting a characterization of the neural activity based on the machine learning model.