Patent Application
20200203011
The disclosure relates to a virtual consultation method and an electronic device.
Generally, patients with chronic conditions need home care and condition tracking after discharge. Nevertheless, a doctor cannot track daily physiological information of a patent. Taking a patient with heart failure for example, the patient's doctor cannot track daily heart rate changes of the patient. In addition, the doctor cannot know the subjective symptoms of the patient face to face. Further, the doctor cannot evaluate whether the patient suffers from an adverse drug effect or other side effects caused by a new drug after a medication change, and the doctor cannot know whether the patient has taken medicine on a regular basis either.
The disclosure provides a virtual consultation method and an electronic device capable of simulating a question asked by a doctor for a patient during consultation.
The disclosure provides a virtual consultation method configured for an electronic device, and the method includes the following steps. Physiological information is obtained through sensing a user by a sensing device is received. The physiological information is analyzed to obtain an analysis result. Weights of a plurality of questions are adjusted according to the analysis result and at least one first question applicable to the user and an order of the at least one first question are determined according to the weights. The at least one first question is outputted according to the order to simulate a question asked by a doctor for the user during consultation.
The disclosure further provides an electronic device including an output/input device and a processor. The processor is coupled to the output/input device. The output/input device receives physiological information obtained through sensing a user by a sensing device. The processor analyzes the physiological information to obtain an analysis result. The processor adjusts weights of a plurality of questions according to the analysis result and determines at least one first question applicable to the user and an order of the at least one first question according to the weights. The output/input device outputs the at least one first question according to the order to simulate a question asked by a doctor for the user during consultation.
To sum up, in the virtual consultation method and the electronic device provided by the disclosure, the physiological information may be obtained through an existing remote instrument, and the questions applicable to the user may be outputted according to the analysis result of the physiological information, so that the question raised by the doctor for the user during consultation may be simulated.
With reference to
The electronic device 100 may include a processor 10 an input/output circuit 12, and a storage device (not shown). The input/output circuit 12 and the storage device are coupled to the processor 10. In addition, the electronic device 100 may also include other more elements, such as a communication chip, which is not limited herein.
The processor 10 may be a central processing unit (CPU) or other programmable microprocessor for general or special use, a digital signal processor (DSP), a programmable controller, an application specific integrated circuit (ASIC), or any other similar devices or a combination of the foregoing devices.
The input/output circuit 12 is a circuit mainly configured for receiving a signal or a file and may transmit the received signal or file to the processor 10. In addition, the input/output circuit 12 may also be configured for receiving a signal or a file generated by the processor 10 and outputs such signal or file to other interfaces or devices.
The storage device may be a fixed or a movable random access memory (RAM) in any form, a read-only memory (ROM), a flash memory, a hard disk drive (HDD), a solid state drive (SSD), any other similar devices, or a combination of the foregoing devices.
In this embodiment, a plurality of program code snippets are stored in the storage device of the electronic device 100, and the program code snippets are executed by the processor 10 of the electronic device 100 after being installed. For instance, the storage device of the electronic device 100 includes a plurality of modules. Operations applied in the electronic device 100 are respectively executed through the modules, and each of the modules is formed by one or plural program code snippets. Nevertheless, the disclosure is not limited thereto, and the operations may also be implemented through using other hardware forms.
With reference to
Next, the processor 10 inputs the physiological information in a model to analyze the received physiological information to obtain an analysis result (step S203). The processor adjusts weights of a plurality of questions according to the analysis result through the model and determines a question (aka a first question) applicable to the user and an order of the first question according to the weights (step S205). That is, in the disclosure, questions and content of the questions for the user are dynamically adjusted, and that a personalized question group is accordingly generated.
In particular, in step S205, the processor 10 may also obtain answer content (aka second answer content) previously filled in by the user and corresponding to another question (aka a second question) and an order of the second question through the output/input device 12. The processor 10 adjusts the weights of the plurality of questions according to the analysis result, the second answer content, and the order of the second question and determines the first question applicable to the user and the order of the first question according to the weights.
With reference to
Next, the processor 10 performs a feature capturing operation on the training data to obtain a plurality of features (step S303). The feature capturing operation may be obtained through a conventional machine learning method, which is not provided herein. The processor 10 normalizes the plurality of features and other data of the user to obtain normalization data (aka first normalization data) (step S305). The processor 10 normalizes the first normalization data and history data of the user to obtain normalization data (aka second normalization data) (step S307). Finally, the processor 10 trains the model according to the second normalization data to obtain the trained model step (S309). How to train a model may be obtained from the prior art, and related description is not provided herein.
Note that in step S307, the first normalization data and the history data of the user himself/herself are normalized, and in this way, outputted contents of different users may be different. For instance, if a user is assumed to have chronic high blood pressure, the measured blood pressure of such user is usually higher than the normal standard. After step S307 is performed, a range of the regular blood pressure of this user is obtained, and a warning is provided or a risk level is increased only when the measured pressure exceeds this range. As such, the user is prevented from being frequently provided with a warning as the measured blood pressure of the user is usually high.
With reference to
In addition, when receiving the first question, the user may answer (or respond to) the first question. The processor 10 may obtain answer content (aka first answer content) from the user corresponding to the first question through the output/input device 12. The processor 10 then determines the risk level and the like corresponding to a physiological condition according to the first answer content. The output/input device 12 outputs a corresponding output message according to the risk level. For instance, when the first question is related to heart disease, the physiological condition may include blood pressure or heart rate, and the risk level may be configured to be an indicator presenting blood pressure or heart rate. Taking blood pressure for example, when the risk level of blood pressure is relatively high (e.g., higher than a threshold), an output message corresponding to the blood pressure may be presented in red to act as a warning.
On the contrary, when the risk level of the blood pressure is relatively low (e.g., lower than the threshold), the output message corresponding to the blood pressure may be presented in green. In an embodiment, when the output message includes various physiological conditions, the physiological conditions may be sorted according to high and low levels of risk levels of the physiological conditions. For instance, the risk levels may be ranked from high to low to facilitate observation made by the user or the doctor.
In particular, an integrated report for a patent or an extension goal may be responded through a semi-automatic, manual, or full automatic manner according to the output message. In the semi-automatic manner, the medical staff may verify correctness of content of the output message including “full smart mode interpretation” and “overall report and details” provided by the model (aka artificial smart engine) and provides “integrated disease interpretation” to the patent. At this stage, the model may also perform learning according to feedbacks provided by the medical staff.
In the manual manner, the medical staff may provide the “integrated disease interpretation” to the patent with reference to the output message including the “full smart mode interpretation” and “overall report and details” provided by the model and may actively intervene adequately. For instance, the medical staff may learn about medication use of the patent and provide suggestions about lifestyle adjustment, actively inform of an early return visit, and adjust medication according to drug reactions and side effects and so on. At this stage, the model may also perform learning according to feedbacks provided by the medical staff.
In the full automatic manner, the output message of the “full smart mode interpretation” and “overall report and details” as well as the “integrated disease interpretation” are automatically provided to the user through the model instead of the medical staff. In addition, the model may provide an early warning of a full smart mode, and such warning may further be provided to: a relatively healthy person having a higher risk, such as a family member of the patent (owing to family history reason). For the general public, preventive tracking may also be performed.
In view of the foregoing, in the virtual consultation method and the electronic device provided by the disclosure, the physiological information may be obtained through an existing remote instrument, and the questions applicable to the user may be outputted according to the analysis result of the physiological information, so that the question raised by the doctor for the user during consultation may be simulated. Moreover, early prevention of diseases, lowered fatality rate, decreased nursing care costs, and other effects may also be generated through the method provided by the disclosure.
This application claims the priority benefit of U.S. provisional application Ser. No. 62/760,044, filed on Nov. 13, 2018. The entirety of the above-mentioned patent application is hereby incorporated by reference herein and made a part of specification.
| Number | Date | Country | |
|---|---|---|---|
| 62760044 | Nov 2018 | US |
