Intelligent method and intelligent system for processing physiological data
An intelligent method and an intelligent system for processing physiological data are provided. The method is operated in a cloud system for implementing a health cloud. A database of the intelligent system stores personalized physiological data collected from various user-end devices, and physiological data transmitted from various medical data sources. A big data is therefore formed. The intelligent system receives continuous physiological data and non-continuous physiological data from each of the user-end devices by a communication circuit. The intelligent system performs big-data analysis to establish a physiological prediction model for predicting various physiological states, and therefore determines a physiological status according to the continuous physiological data or the non-continuous physiological data transmitted from each of the user-end devices. A feedback message is accordingly created and sent to a user in real time.
1 . An intelligent system for processing physiological data, the intelligent system comprising:
at least two user-end sensors including a full-time-activated user-end sensor that is used to continuously generate continuous physiological data to be transmitted to a cloud system and an event-activated user-end sensor that is activated in response to a feedback message for generating non-continuous physiological data to be transmitted to the cloud system instantly or for a period of time;
a button acting as an activation interface being disposed on the full-time-activated user-end sensor; and
a database implemented by a storage device, used to store personalized physiological data generated by the at least two user-end sensors that are configured to sense physiological status of a user, so as to form a big data;
wherein the intelligent system performs an intelligent method for processing physiological data comprising operations of:
receiving the continuous physiological data and the non-continuous physiological data transmitted from the at least two user-end sensors, wherein the non-continuous physiological data is generated by the event-activated user-end sensor which is activated by the user and/or by the cloud system; wherein the continuous physiological data includes the physiological data that full-time-activated user-end sensor continuously or periodically uploads to the cloud system and the continuous and non-continuous physiological data form a part of the big data stored in the database;
analyzing the continuous physiological data and the non-continuous physiological data transmitted from the at least two user-end sensors and comparing with a physiological prediction model established by a machine-learning method in the cloud system for predicting various physiological states;
generating the feedback message with respect to the continuous physiological data or the non-continuous physiological data according to the physiological state obtained by analyzing the continuous or non-continuous physiological data transmitted from the at least two user-end sensors; and
transmitting, via a communication network, the feedback message to a communication device of the user according to identification and communication information of each of the at least two user-end sensors, and the cloud system is able to optimize the physiological prediction model when the user responds to the feedback message;
wherein, when the activation interface is triggered to generate an activation signal, the event-activated user-end sensor is activated for sensing the physiological data with a minority of items for a period of time that forms the non-continuous physiological data transmitted to the cloud system, and the physiological data generated by the event-activated user-end sensor is transmitted to the cloud system and is labeled for establishing a correlation between the continuous physiological data that is generated by the full-time-activated user-end sensor and the non-continuous physiological data relating to the feedback message.
2 . The intelligent system according to claim 1 , wherein the full-time-activated user-end sensor or the event-activated user-end sensor is a smart bracelet, a smart necklace, smart clothes, or a fixed physiological sensing sensor worn by the user.
3 . The intelligent system according to claim 1 , wherein the means for machine-learning method utilizes the big data collected in the database to establish and train the physiological prediction model used to predict various physiological states, and the physiological prediction model is optimized by referring to the physiological state of the user corresponding to each piece of the non-continuous physiological data.
4 . The intelligent system according to claim 3 , wherein the operation of analysis is an artificial intelligence that is used to determine one of the physiological states according to various preset physiological states of the physiological prediction model based on the continuous physiological data or the non-continuous physiological data transmitted by the at least two user-end sensors.
5 . An intelligent method for processing physiological data, which is performed in a cloud system that operates an intelligent system comprising at least two user-end sensors including a full-time-activated user-end sensor and an event-activated user-end sensor, an activation interface being disposed on the full-time-activated user-end sensor and a storage-device-implemented database used to store personalized physiological data generated by the at least two user-end sensors which are configured to sense physiological status of a user, wherein the intelligent method comprises:
receiving continuous physiological data and non-continuous physiological data from the at least two user-end sensors having the full-time-activated user-end sensor and the event-activated user-end sensor worn by the user, wherein the continuous physiological data includes the physiological data that full-time-activated user-end sensor continuously or periodically uploads to the cloud system;
analyzing the continuous physiological data or the non-continuous physiological data, and comparing with a physiological prediction model established by a machine-learning method in the cloud system, so as to determine a physiological state;
generating a feedback message with respect to the continuous physiological data or the non-continuous physiological data in response to the physiological state; and
transmitting, via a communication network, the feedback message to a communication device of the user according to identification and communication information corresponding of each of the at least two user-end sensors, and the cloud system is able to optimize the physiological prediction model when the user responds to the feedback message;
wherein the personalized physiological data includes the continuous physiological data and the non-continuous physiological data that form a big data; and the big data stored in the database is then analyzed by the machine-learning method, so as to establish the physiological prediction model used to predict the physiological state;
wherein, when the activation interface is triggered to generate an activation signal, both the full-time-activated user-end sensor and the event-activated user-end sensor are cooperatively activated; wherein, when the activation signal is generated via the activation interface, the event-activated user-end sensor is configured to sense the physiological data with a minority of items for a period of time that forms the non-continuous physiological data transmitted to the cloud system, and the physiological data generated by the event-activated user-end sensor is transmitted to the cloud system and is labeled for establishing a correlation between the continuous physiological data that is generated by the full-time-activated user-end sensor and the non-continuous physiological data relating to the feedback message.
6 . The intelligent method according to claim 5 , wherein the event-activated user-end sensors is activated if the cloud system determines an abnormality based on the continuous physiological data, the full-time-activated user-end sensor is activated cooperatively, and the physiological data generated by both the full-time-activated user-end sensor and the event-activated user-end sensor for the period of time is transmitted to the cloud system.
7 . The intelligent method according to claim 5 , wherein the big data collected in the database is utilized by the machine-learning method to establish and train the physiological prediction model used to predict various physiological states, and the user's state corresponding to each piece of the non-continuous physiological data is referred to for optimizing the physiological prediction model; wherein the means for analysis is an artificial intelligence that is used to determine one of the physiological states according to various preset physiological states of the physiological prediction model based on the continuous physiological data or the non-continuous physiological data transmitted by any of the full-time-activated user-end sensor and the event-activated user-end sensor.
8 . An intelligent system for processing physiological data, the intelligent system comprising:
at least two user-end sensors including a full-time-activated user-end sensor that is a wearable sensor worn on a care recipient and is used to continuously generate continuous physiological data to be transmitted to a cloud system in a full-time operation, and an event-activated user-end sensor that is disposed around the care recipient and is activated in response to a feedback message for generating non-continuous physiological data to be transmitted to the cloud system instantly or for a period of time;
a button acting as an activation interface being disposed on the full-time-activated user-end sensor; and
a database implemented by a storage device, used to store personalized physiological data generated by the at least two user-end sensors that are configured to sense physiological status of a user, so as to form a big data;
wherein the intelligent system performs an intelligent method for processing physiological data comprising operations of:
receiving the continuous physiological data and the non-continuous physiological data transmitted from the at least two user-end sensors, wherein the non-continuous physiological data is generated by the event-activated user-end sensor which is activated by the user and/or by the cloud system; wherein the continuous physiological data includes the physiological data that full-time-activated user-end sensor continuously or periodically uploads to the cloud system and the continuous and non-continuous physiological data form a part of the big data stored in the database;
analyzing the continuous physiological data and the non-continuous physiological data transmitted from the at least two user-end sensors and comparing with a physiological prediction model established by a machine-learning method in the cloud system for predicting various physiological states;
generating the feedback message with respect to the continuous physiological data or the non-continuous physiological data according to the physiological state obtained by analyzing the continuous or non-continuous physiological data transmitted from the at least two user-end sensors; and
transmitting, via a communication network, the feedback message to a communication device of the user according to identification and communication information of each of the at least two user-end sensors, and the cloud system is able to optimize the physiological prediction model when the user responds to the feedback message;
wherein, when the activation interface is triggered to generate an activation signal, the event-activated user-end sensor is activated for sensing the physiological data with a minority of items for a period of time that forms the non-continuous physiological data transmitted to the cloud system, and the physiological data generated by the event-activated user-end sensor is transmitted to the cloud system and is labeled for establishing a correlation between the continuous physiological data that is generated by the full-time-activated user-end sensor and the non-continuous physiological data relating to the feedback message.