Sensing system with features for determining and predicting brain age and other electrophysiological metrics of a subject

Inventors

{hacek over (S)}arlija, Marko • Comerford, III, Michael Kenneth • Crow, Karen

Assignees

Neurogeneces Inc

Abstract

Some systems, devices and methods detailed herein provide a system for use in determining metrics of a subject. The system can provide, as an output, a function-metric value determined based on a defined relationship between physiological measures and a chronological age.

Core Innovation

The invention relates to a system for use in determining metrics of a subject by receiving physiological measures of brain activity recorded at least partly while the subject is asleep, and receiving demographic data including a chronological age for the subject when the physiological measures were recorded. The system generates segmented training-data specifying a plurality of epochs of time and data for the subject in each epoch, and then generates sleep-structure features for the subject.

Using the segmented training-data, the system selects a subset of the sleep-structure features as selected features, and generates one or more function-metric classifiers. The function-metric classifier trains a model that defines at least one relationship between the physiological measures and the chronological age, and receives new physiological measures as input and provides a function-metric value as output based on the defined relationship.

In one form, the function-metric value comprises a variance-from-chronological-age value that includes an indication of the subject’s brain function compared to the subject’s expected brain function based on the subject’s chronological age. In another form, the system estimates a predicted function-metric for the subject to represent a measure of predicted future physiological measures based on an expected change to the brain activity due to advancing in chronological age.

Claims Coverage

The provided content includes three independent claims covering epoch-based sleep data segmentation, sleep-structure feature generation and selection, and function-metric classifiers tied to chronological age. The claims also cover output of a variance-from-chronological-age value and a predicted function-metric representing expected future physiological measures.

The inventive coverage centers on epoch-based sleep data, sleep-structure features, feature selection, and trained function-metric classifiers that output either a variance-from-chronological-age value or a predicted function-metric tied to advancing age.

Stated Advantages

Documented Applications

No documented applications found