Back to results
US 2019/0336824 A1Published application
System and method for predictive health monitoring
AssigneeBlue Goji LLC
InventorColeman Fung
Life of this patent
Estimating the term…
Abstract
A system and method for predictive health monitoring using neural networks, comprising a wearable device with biometric sensors, a database containing data from multiple users across many categories of health-related factors, a first set of neural networks trained on the database that makes health predictions based on a single health factor, and a second neural network that makes health predictions based on a combination of the predictions made by the first set of neural networks.
Claims14
A system for predictive health monitoring, comprising:
- a cloud-based health prediction engine comprising:
- a plurality of first-stage neural networks, each configured to make a first health prediction based on a health-related factor;
- a second-stage neural network, configured to make a second health prediction based on a combination of the first health predictions from at least two of the plurality of first-stage neural networks;
- a data storage device configured to store a history of biometric data and a history of health predictions for a user of a wearable biometric monitoring and feedback device;
- a network-connected server comprising a memory, a processor, and a plurality of programming instructions, wherein the programming instructions, when operating on the processor, cause the network-connected server to:
- receive biometric data from a wearable biometric monitoring and feedback device for the user;
- retrieve the history of biometric data and the history of health predictions for the user from the data storage device;
- process the biometric data, history of biometric data, and the history of health predictions through at least two of the plurality of first-stage neural networks;
- receive the first health prediction from each first-stage neural network through which the biometric data was processed;
- process the first health predictions received through a second-stage neural network;
- receive the second health prediction from the second-stage neural network;
- send the second health prediction to the wearable biometric monitoring and feedback device, and
- a wearable biometric monitoring and feedback device comprising:
- a plurality of sensors for gathering biometric data from the user of the wearable biometric monitoring and feedback device;
- a network device configured to connect to the cloud-based health prediction system;
- a screen for providing feedback to the user;
- and
- a computing device comprising a memory, a processor, and a plurality of programming instructions, wherein the programming instructions, when operating on the processor, cause the computing device to:
- obtain biometric data from at least two of the plurality of sensors for the user of the wearable biometric monitoring and feedback device;
- send the biometric data obtained to the cloud-based health prediction engine using the network device;
- receive a second health prediction from a cloud-based health prediction engine;
- and
- display the second health prediction to the user.
20 more elements
A method for predictive health monitoring, comprising the steps of:
- obtaining biometric data for a user of a wearable biometric monitoring and feedback device;
- retrieving a history of biometric data and a history of health predictions for the user from a data storage device;
- processing the biometric data, history of biometric data, and the history of health predictions through at least two of a plurality of first-stage neural networks, the plurality of first-stage neural networks, each configured to make a first health prediction based on a separate health-related factor;
- receiving a first health prediction from each first-stage neural network through which the biometric data was processed;
- processing the first health predictions received through a second-stage neural network, the second-stage neural network configured to make a second health prediction based on a combination of the first health predictions from at least two of the plurality of first-stage neural networks;
- receiving the second health prediction from the second-stage neural network;
- and
- displaying the second health prediction to the user of the wearable biometric monitoring and feedback device.
5 more elements
Dependencies are inferred from the claim text.