Deep learning architecture for cognitive examination subscore trajectory prediction in Alzheimer's disease
US10898125B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Jan 21, 2020 |
| Grant date | Jan 26, 2021 |
| Priority date | — |
| Expiry date | Jan 21, 2040 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/30016
- WIPO fieldMedical technology
- WIPO sectorInstruments
Abstract
A method of predicting progression of a cognitive state in a subject is disclosed including obtaining a neuroimage of the subject, acquiring a data sample from the neuroimage, selecting a time at which to predict progression of the cognitive state, and performing a calculation on the data sample by a transformation function to determine data associated with multiple cognitive metrics. A method of evaluating a cognitive state of a subject is also disclosed including providing a convolutional neural network, training the convolutional neural network with reference data to construct a transformation function, using the transformation function to predict multiple cognitive metrics from a subject data sample and a selected time, and determining the cognitive state of the subject from the predicted cognitive metrics. A cognitive evaluation system is also disclosed including a memory, a processor, and a cognitive state prediction component configured to program the processor with a transformation function.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.