Method and apparatus for determining dementia risk factors using deep learning

Patent metadata

Field	Value
Publication number	US-12367979-B2
Application number	US-202117381564-A
Country	US
Kind code	B2
Filing date	Jul 21, 2021
Priority date	Jul 21, 2021
Publication date	Jul 22, 2025
Grant date	Jul 22, 2025

Abstract

There is provided a method for determining dementia risk factors by a server using deep learning. In this instance, the method for determining dementia risk factors includes acquiring biometric information from each subject corresponding to a first control group through a wearable device, acquiring measurement information for each subject corresponding to the first control group, deriving a first dementia risk factor based on the biometric information and the measurement information for each subject, and deriving a second dementia risk factor related to the first dementia risk factor via deep learning performed based on the biometric information related to the first dementia risk factor and control group information.

First claim

The invention claimed is:

1. A system for determining dementia risk factors, comprising:
a server comprising an artificial neural network (ANN), a transmitter of the server, a receiver of the server, a processor of the server, a memory of the server communicatively coupled to the processor of the server and storing instructions operable when executed by the processor of the server to perform a predetermined function and automatically determining the dementia risk factors with a biometric information and a dementia related information, wherein
the dementia related information comprises a first dementia related information, a second dementia related information, a third dementia related information and a fourth dementia related information, and
the ANN is implemented as an application specific integrated circuit (ASIC);
a wearable device comprising a transmitter of the wearable device, a receiver of the wearable device, an acceleration sensor, a motion sensor, a processor of the wearable device, a memory of the wearable device communicatively coupled to the processor of the wearable device and storing instructions operable when executed by the processor of the wearable device to perform a predetermined function, automatically measuring the biometric information including consumed calories, a spent activity time, a sleep duration, a type of sleep for each sleep time, a heart rate, a number of steps, a walking distance and a number of exercises and communicating with a plurality of Internet of Things (IoT) devices;
the plurality of Internet of Things (IoT) devices, each of the plurality of IoT devices comprising a processor of the IoT devices, a memory of the IoT devices communicatively coupled to the processor of the IoT devices and storing instructions operable when executed by the processor of the IoT devices to perform a predetermined function, a transmitter and a receiver and automatically communicate with the server and the wearable device, wherein:
the plurality of IoT devices comprises a first IoT device, a second IoT device and a third IoT device;
the first IoT device is a device that is fixed at a predetermined location, operated based on a IoT cloud, automatically receives a signal from the wearable device and transmits data to the server and the second and third IoT devices through the IoT cloud;
the second IoT device is a device with a power of a predetermined level which is lower than a power of the first and third IoT devices; and
the third IoT device is a device that is installed in an indoor space and automatically measures a temperature, a humidity, an air pollution level and an illuminance of an indoor environment of a wearer of the wearable device, wherein the third IoT device further automatically measures an information of a start time and an end time of a workday and the sleep duration of the wearer of the wearable device via a connection to the wearable device and automatically transmits the information of the start time and the end time to the server;
a computed tomography (CT) scanner measuring the first dementia related information;
a magnetic resonance imaging (MRI) scanner measuring the second dementia related information;
an electroencephalogram (EEG) recorder measuring the third dementia related information;
an amyloid positron emission tomography (PET) scanner measuring a fourth dementia related information; and
a blood biomarker measuring the fifth dementia related information.

Assignees

• The Catholic Univ Of Korea Industry—Academic Cooperation Foundation
• Gil Medical Ct
• Univ Inha Res & Business Found
• Wang Min Jeong
• Catholic Univ Korea Ind Academic Cooperation Foundation

Inventors

• Yang Dong Won
• Kim Sang Yun
• Park Kee Hyung
• Jeong Jee Hyang
• Choi Seong Hye
• Hong yun jeong
• Wang Min Jeong

Classifications

• G06N3/08

  Learning methods · CPC title

• G16H10/60

  for patient-specific data, e.g. for electronic patient records · CPC title

• G16H15/00

  ICT specially adapted for medical reports, e.g. generation or transmission thereof · CPC title

• A61B5/4088

  Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia · CPC title

• A61B2560/0242

  adapted to measure environmental factors, e.g. temperature, pollution · CPC title