CN-121999968-A - Health care recommendation method and system using rehabilitation medical materials

CN121999968ACN 121999968 ACN121999968 ACN 121999968ACN-121999968-A

Abstract

A health care recommendation method and system using a rehabilitation medical material, wherein the health care recommendation method is executed by a processing device and comprises the steps of obtaining a plurality of pieces of original sensing data from a plurality of sensing devices, marking one of the plurality of pieces of original sensing data according to a plurality of personalized features to generate marked data, inputting one of the plurality of pieces of original sensing data into a pre-training prediction model to generate predicted data, fusing the marked data and the predicted data to generate fused data, extracting multi-dimensional feature data from the plurality of pieces of original sensing data, inputting the fused data and the multi-dimensional feature data into the pre-training inference model to generate a motion achievement index, and generating a health care plan according to the motion achievement index.

Inventors

WU ZHILIN
ZHONG YUNCHENG
CHEN ZHIYING
LIN QIANDE

Assignees

财团法人资讯工业策进会

Dates

Publication Date: 20260508
Application Date: 20241121
Priority Date: 20241106

Claims (16)

1. A health care recommendation method performed by a processing device, comprising: Fetching a plurality of raw sensing data from a plurality of sensing devices; Marking one of the plurality of raw sensed data according to a plurality of personalized features to produce marked data; inputting one of the plurality of raw sensed data into a pre-trained predictive model to generate predictive data; performing data fusion on the marking data and the prediction data to generate fusion data; extracting multi-dimensional feature data from the plurality of raw sensed data; Inputting the fusion data and the multidimensional feature data into a pre-training inference model to generate a sports performance index, and And generating a health care plan according to the athletic performance index.
2. The health care recommendation method of claim 1, wherein the plurality of raw sensed data includes a movement time and a movement trajectory, and the health care recommendation method further includes: calculating an acceleration change using the motion time and the motion trajectory, and And performing action fluency analysis by using the acceleration change to generate fluency data.
3. The health care recommendation method of claim 2, wherein data fusing the tagging data and the predictive data to generate the fused data further comprises: and carrying out data fusion on the marking data, the prediction data and the fluency data to generate fusion data.
4. The health care recommendation method of claim 2, wherein utilizing the acceleration changes for action fluency analysis to generate the fluency data comprises: obtaining the motion type corresponding to the motion trail and And selecting one from a plurality of motion fluency analysis algorithms according to the motion type to perform the motion fluency analysis.
5. The health care recommendation method as claimed in claim 1, further comprising: Acquiring feedback data of the user, and And adjusting the threshold setting of marking according to the personalized features according to the feedback data of the user.
6. The health care recommendation method as claimed in claim 1, further comprising: Acquiring first historical sensing data and user feedback data corresponding to the first historical sensing data; Marking the first history sensing data according to threshold setting and a plurality of preset action characteristics to generate first history marking data; adjusting the threshold setting according to the user feedback data and the first history flag data; Marking the plurality of second history sensing data according to the adjusted threshold setting and the plurality of preset motion characteristics to generate a plurality of second history marking data, and Generating a normalcy model using the second plurality of historical tag data, wherein the normalcy model is indicative of the plurality of personalized features.
7. The health care recommendation method as claimed in claim 1, further comprising: acquiring a plurality of historical sensing data from the plurality of sensing devices; generating a plurality of training data using the plurality of historical sense data; Training with the plurality of training data to generate a multi-dimensional inference model, and The multi-dimensional inference model is approximated and compressed to generate the pre-trained inference model.
8. The health care recommendation method as claimed in claim 1, further comprising: acquiring a plurality of historical sensing data from the plurality of sensing devices; Tagging one of the plurality of historical sensed data according to the plurality of personalized features to generate historical tagging data; inputting one of the plurality of historical sensed data into the pre-trained predictive model to generate historical predicted data; performing data fusion on the history marking data and the history prediction data to generate history fusion data; extracting historical multi-dimensional feature data from the plurality of historical sense data; taking the history fusion data and the history multidimensional feature data as one of a plurality of training data, and The pre-training inference model is generated using the plurality of training data.
9. A health care recommendation system, comprising: a plurality of sensing devices for acquiring a plurality of original sensing data, and The processing device is connected with the plurality of sensing devices and is used for marking one of the plurality of original sensing data according to a plurality of personalized features to generate marked data, inputting one of the plurality of original sensing data into a pre-training prediction model to generate predicted data, carrying out data fusion on the marked data and the predicted data to generate fusion data, extracting multi-dimensional feature data from the plurality of original sensing data, inputting the fusion data and the multi-dimensional feature data into a pre-training inference model to generate a motion achievement index, and generating a health care plan according to the motion achievement index.
10. The health care recommendation system of claim 9, wherein the plurality of raw sensing data includes a movement time and a movement trajectory, and the processing device is further configured to calculate an acceleration variation using the movement time and the movement trajectory, perform a fluency analysis using the acceleration variation to generate fluency data, and fuse the marking data, the prediction data, and the fluency data to generate the fused data.
11. The health care recommendation system of claim 10 wherein the processing device is further configured to perform the data fusion of the marking data, the prediction data, and the fluency data to generate the fused data.
12. The health care recommendation system of claim 10, wherein the processing device is configured to obtain a type of movement corresponding to the movement trajectory and select one from a plurality of movement smoothness analysis algorithms according to the type of movement to perform the movement smoothness analysis.
13. The health care recommendation system of claim 9, wherein the processing device is further configured to obtain user feedback data and adjust a threshold setting for marking based on the plurality of personalized features based on the user feedback data.
14. The health care recommendation system of claim 9, wherein the processing device is further configured to: Acquiring first historical sensing data and user feedback data corresponding to the first historical sensing data; Marking the first history sensing data according to threshold setting and a plurality of preset action characteristics to generate first history marking data; adjusting the threshold setting according to the user feedback data and the first history flag data; Marking the plurality of second history sensing data according to the adjusted threshold setting and the plurality of preset motion characteristics to generate a plurality of second history marking data, and Generating a normalcy model using the second plurality of historical tag data, wherein the normalcy model is indicative of the plurality of personalized features.
15. The health care recommendation system of claim 9, wherein the processing device is further configured to: acquiring a plurality of historical sensing data from the plurality of sensing devices; generating a plurality of training data using the plurality of historical sense data; Training with the plurality of training data to generate a multi-dimensional inference model, and The multi-dimensional inference model is approximated and compressed to generate the pre-trained inference model.
16. The health care recommendation system of claim 9, wherein the processing device is further configured to: acquiring a plurality of historical sensing data from the plurality of sensing devices; Tagging one of the plurality of historical sensed data according to the plurality of personalized features to generate historical tagging data; inputting one of the plurality of historical sensed data into the pre-trained predictive model to generate historical predicted data; performing data fusion on the history marking data and the history prediction data to generate history fusion data; extracting historical multi-dimensional feature data from the plurality of historical sense data; taking the history fusion data and the history multidimensional feature data as one of a plurality of training data, and The pre-training inference model is generated using the plurality of training data.

Description

Health care recommendation method and system using rehabilitation medical materials Technical Field The invention relates to a health care recommendation method and a system. Background Population aging has become a global trend and is rapidly advancing, so how to effectively care for the population over the aging in the future is a great challenge for society. In the face of the arrival of advanced society, caregivers need to integrate and use resource communication technology products and services to solve the problem of manpower shortage and improve the care quality of the long people. In the current rehabilitation care process, the operation evaluation of rehabilitation medical materials is mainly performed according to the operation times and time, or by adopting a general standard normal model. However, these methods often ignore the physiological condition, the operation process and the personal health history data of the individual, and cannot perform personalized performance evaluation according to the multidimensional data, which limits the accuracy and pertinence of the rehabilitation effect. Disclosure of Invention In view of the above, the present invention provides a health care recommendation method and system for solving the above problems. A health care recommendation method according to an embodiment of the invention is performed by a processing device and includes obtaining a plurality of pieces of raw sensing data from a plurality of sensing devices, tagging one of the plurality of pieces of raw sensing data according to a plurality of personalized features to generate tagged data, inputting one of the plurality of pieces of raw sensing data into a pre-trained predictive model to generate predicted data, data fusing the tagged data and the predicted data to generate fused data, extracting multi-dimensional feature data from the plurality of pieces of raw sensing data, inputting the fused data and the multi-dimensional feature data into a pre-trained inference model to generate a athletic performance index, and generating a health care plan according to the athletic performance index. According to an embodiment of the invention, a health care recommendation system comprises a plurality of sensing devices and a processing device. The plurality of sensing devices are used for acquiring a plurality of original sensing data. The processing device is connected with the plurality of sensing devices and is used for marking one of the plurality of original sensing data according to a plurality of personalized features to generate marked data, inputting the one of the plurality of original sensing data into a pre-training prediction model to generate predicted data, carrying out data fusion on the marked data and the predicted data to generate fusion data, extracting multi-dimensional feature data from the plurality of original sensing data, inputting the fusion data and the multi-dimensional feature data into a pre-training inference model to generate a motion achievement index, and generating a health care plan according to the motion achievement index. In summary, according to the health care recommendation method and system of one or more embodiments of the present invention, the sensed data can be personalized labeled, the feature extraction of multiple data dimensions and the processing of multiple model data, and the personalized athletic performance index evaluation is performed by using the data, so as to provide a user health care scheme, thereby solving the problem of manpower shortage of the care organization and improving the care quality. The foregoing description of the disclosure and the following description of embodiments are presented to illustrate and explain the spirit and principles of the invention and to provide a further explanation of the invention as claimed. Drawings Fig. 1 is a block diagram of a health care recommendation system according to an embodiment of the invention. Fig. 2 is a flowchart of a health care recommendation method according to an embodiment of the invention. Fig. 3 is a flowchart illustrating a method for analyzing the smoothness of actions in a health care recommendation method according to an embodiment of the invention. Fig. 4 is a flowchart illustrating a method for establishing personalized features in a healthcare recommendation method according to an embodiment of the invention. Fig. 5 is a flowchart of a method for obtaining a pre-training inference model in a health care recommendation method according to an embodiment of the invention. Fig. 6 is a flowchart of a method for obtaining a pre-training inference model in a health care recommendation method according to another embodiment of the invention. Detailed Description The detailed features and advantages of the present invention will be readily apparent to those skilled in the art from the following detailed description, claims, and drawings that are provided herein. The following examp