TW202303634A - Sarcopenia assessment system and method and model establishing method - Google Patents

Abstract

The present invention relates to a sarcopenia assessment method, comprising: providing a sarcopenia self-inspection for a user and acquiring an inspection result; uploading and feeding the inspection result to a sarcopenia assessment model established on the basis of a machine learning classifier; and executing the sarcopenia assessment model in a remote end to performing a sarcopenia assessment, to assess whether the user belongs to a high risk group of sarcopenia or not.

Description

Sarcopenia assessment system and method and assessment model establishment method

The present invention relates to a sarcopenia assessment system and method, and a method for establishing a sarcopenia assessment model based on artificial intelligence machine learning technology, in particular, it can provide users with simple operation of their own user equipment to perform sarcopenia autonomously Systems and methods for risk assessment.

Sarcopenia refers to the continuous and general decrease in skeletal muscle weight and function, which may lead to disability, decreased quality of life, even inability to take care of oneself and increased risk of death. Studies have pointed out that human skeletal muscle will grow with age And reduce, over the age of 40, the muscle mass will be lost at a rate of 8% per decade, and after the age of 70, the loss will be accelerated at a rate of 15% per decade. If it is reduced by 1%, the muscle mass and muscle strength may be lost by as much as 50% by the age of 80.

The word sarcopenia first originated in English in 1989. Irwin Rosenberg proposed to combine the Greek roots representing flesh (sarx) and lack (penia) into the word "sarcopenia" to describe the occurrence of skeletal disorders due to aging. The phenomenon of loss of muscle mass, and the definition of sarcopenia, generally adopts the definition of sarcopenia by the European Working Group on Sarcopenia (EWGSOP) in 2010, which is: "progressive loss of muscle mass and reduction of muscle strength or physical activity , and divide sarcopenia into pre-sarcopenia, sarcopenia and severe sarcopenia".

However, due to the different cut-off points for the definition of sarcopenia, there are still different definitions of sarcopenia. For example, Baumgartner et al. defined muscle mass index (appendicular muscle mass/height 2) as two standard deviations lower than the average value of young people as Sarcopenia, and then Newman, Janssen et al. also used a similar method, but corrected by skeletal muscle, and then defined sarcopenia by percentage or standard deviation. The International Sarcopenia Working Group (IWGS) proposed that in addition to the aforementioned reduction in muscle mass and strength, low physiological performance should also be added. The Society for Sarcopenia, Cachexia, and Wasting Disease (SCWD) believes that in addition to the reduction of skeletal muscle, activity limitation, such as a reduction in walking speed, is required. The Special Interest Group (SIG) on cachexia-anorexia in chronic wasting diseases believes that sarcopenia is not limited to the elderly, and symptoms include a decrease in muscle mass and strength.

The causes of sarcopenia are thought to be multifactorial, including hormonal changes due to age and aging, neuronal degeneration, decreased number of neuromuscular junctions, insufficient blood vitamin D, nutritional deficiencies, reduced activity levels, reduced levels of physical activity, static Sedentary lifestyle and increased oxidative stress in the body will all contribute to the occurrence of sarcopenia. Gender, history of chronic diseases, high or low body mass index (BMI), previous falls, poor physical function, lack of exercise, long-term drug use (such as steroids), insufficient calorie or protein intake, gastrointestinal diseases, kidney diseases, Retirement, low education, bed rest, and hospitalization all accelerate muscle loss and are considered high-risk groups for sarcopenia.

Sarcopenia will lead to adverse prognosis such as the risk of falls and the chance of fracture, physical activity dysfunction, frailty and disability of the elderly, and even increase the risk of death in severe cases. The cognition and definition are still unclear, and there are even many elderly people who are diagnosed with sarcopenia only after functional impairment has occurred. In particular, through the National Nutrition Survey conducted in the United States from 1999 to 2010, it was found that the prevalence of sarcopenia in adults over the age of 60 As high as 29.9%, and the skeletal muscle mass will gradually lose with age, and the changes in body composition due to aging will lead to a decline in muscle strength or activity.

But on the other hand, with the development and advancement of information and communication technology (ICT), the World Health Organization (WHO) has proposed the eHealth initiative in 2010, believing that all applications of ICT technology in health-related fields aim to promote personal or community health Both of them can be called eHealth. According to the e-health global survey report released in 2016, more than half of the WHO member states have formulated e-health policies. The so-called e-health refers to the use of communication technologies in the field of health, such as applications (App) is one such application.

Therefore, with the development of technology and the times, it is necessary to build an evaluation model for sarcopenia based on ITC technology, combined with artificial intelligence and machine learning technology, and through a very simple application that everyone can operate or The browser is a system or model that is provided to the general public to self-assess whether they belong to the high-risk group of sarcopenia.

For this reason, in view of the shortcomings in the conventional technology, the inventor has tried and researched carefully, and with a persistent spirit, he finally conceived the "sarcopenia evaluation system and method and evaluation model establishment method" in this case, which can overcome The above-mentioned shortcoming, the following is a brief description of the present invention.

In view of the deficiencies of conventional technologies, the present invention proposes a sarcopenia assessment system and method, and a method for establishing a sarcopenia assessment model based on artificial intelligence machine learning technology. A system and method for easily performing sarcopenia risk assessment by implementing the sarcopenia assessment model at the end.

Accordingly, the present invention proposes a method for evaluating sarcopenia, which includes: providing users with self-detection of sarcopenia through a front-end program, and obtaining the detection result; uploading the detection result to A sarcopenia assessment model based on a machine learning classifier; and executing the sarcopenia assessment model remotely to perform sarcopenia assessment, so as to assess whether the user belongs to a high-risk group for sarcopenia based on the detection result .

The present invention further proposes a sarcopenia assessment system, which includes: a system server, which is installed with a sarcopenia assessment smart platform including a sarcopenia assessment model built based on a machine learning classifier; and a user device, which It is configured separately from the server of the system and connected by communication, and the front-end program of the smart platform for sarcopenia evaluation is installed, so as to implement autonomous detection of sarcopenia to the user through a shortcut interface, and obtain the detection results, wherein the user's device The detection result is uploaded and input into the sarcopenia evaluation model, so as to automatically evaluate whether the user belongs to a high-risk group of sarcopenia according to the detection result.

The present invention further proposes a method for establishing a sarcopenia assessment model, which includes: implementing the sarcopenia action questionnaire survey program, providing a quick interface to multiple respondents through the front-end program to conduct the sarcopenia action questionnaire survey, and obtaining the sarcopenia action questionnaire Survey results; selectively implement labeling procedures, label the survey results of sarcopenia according to predefined rules, classify the survey results of sarcopenia into at least two types of results, and obtain a set of labeled data; and The survey results or the set of labeled data are input into a machine learning classifier to train the machine learning classifier to establish a sarcopenia assessment model. .

The above summary of the invention is intended to provide a simplified summary of the disclosure to enable readers to have a basic understanding of the disclosure. This summary of the invention is not intended to disclose a complete description of the invention, and is not intended to point out important/key elements or components of the embodiments of the invention. define the scope of the invention.

10: Sarcopenia assessment system of the present invention

100: user equipment

101: Processor unit

103: Wireless radio frequency communication module

105:Non-volatile memory unit

107:Touch screen unit

109: Apps

111:Shortcut operation program components

110:Mobile device

120: desktop computer

125: Notebook computer

130:Smartphone

135: Tablet device

150: user

200: System server

250: Sarcopenia database

260: Internet

310: Self-test instruction interface

312: Detect a shortcut key

314: Detect the second shortcut key

320: Test a self-test interface

321: DAX numerical shortcut input field

323:Shortcut input fields around the calf

325: Muscle mass shortcut input field

327: Mobility shortcut input field

329: Execute detection shortcut key

330: Sarcopenia assessment result interface

500: sarcopenia assessment method of the present invention

501-509: Implementation steps

Figure 1 shows a schematic diagram of the system architecture of the sarcopenia assessment system of the present invention;

Figure 2 discloses a view of the hardware network equipment architecture of the system server and user equipment included in the sarcopenia assessment system of the present invention;

Figure 3 reveals the distribution of data features revealed by the unsupervised machine learning classifier of the present invention extracted from the raw data of calf circumference lower than the normal value;

Figure 4 reveals the data feature distribution map revealed by the unsupervised machine learning classifier of the present invention to extract muscle strength (hand grip strength) lower than the normal value raw data;

Figure 5 reveals the distribution of data characteristics revealed by the unsupervised machine learning classifier of the present invention, whose extraction power is lower than the normal value raw data;

Fig. 6 shows a schematic diagram of the sarcopenia self-test instruction interface displayed on the user equipment by the sarcopenia assessment system of the present invention;

Fig. 7 discloses a schematic diagram of a test-self-test interface displayed on a user device by the sarcopenia assessment system of the present invention;

Fig. 8 discloses a schematic diagram of the sarcopenia assessment result interface displayed on the user equipment by the sarcopenia assessment system of the present invention; and

Fig. 9 discloses a flow chart of the implementation steps of the sarcopenia assessment method of the present invention.

The present invention can be fully understood by the following examples, so that those skilled in the art can complete it, but the implementation of the present invention can not be limited by the following examples of implementation; the drawings of the present invention are not limited No limitation on size, dimension and scale is included, and the size, dimension and scale of the present invention are not limited by the drawings of the present invention during the actual implementation.

The word "preferred" in this article is non-exclusive and should be understood as "preferably but not limited to". order, the scope of the present invention should be determined only by the appended claims and their equivalents, not by the examples illustrated in the implementation; when the term "comprising" and its variations appear in the specification and claims, it is An open-ended term without a restrictive meaning that does not exclude other features or steps.

Figure 1 discloses a schematic diagram of the system architecture of the sarcopenia assessment system of the present invention; Figure 2 discloses a view of the hardware network equipment architecture of the system server and user equipment included in the sarcopenia assessment system of the present invention; the present invention The sarcopenia assessment system 10 includes a system server 200 configured at the remote end (server end) using a client-server model, and a system server 200 configured at the front end (server end) client end) multiple user equipment (user equipment, UE) 100, the user equipment 100 is preferably a mobile device 110, a desktop computer 120, a notebook computer 125, a smart phone 130 or a tablet device 135, etc., the system server A communication link is established between the device 200 and multiple user equipments 100 via the Internet 260, so as to perform two-way communication and data exchange. The user equipment 100 includes hardware units such as a processor unit 101, a wireless RF communication module (wireless RF communication module) 103, a non-volatile memory unit (flash memory) 105, and a touch screen unit 107, which are electrically connected to each other, and The application program 109 and software components such as the shortcut operation program component 111 included in the application program 109 , the application program 109 as a front-end program is installed on the user equipment 100 and loaded and executed through the processor unit 101 .

When the user 150 starts the application program 109 on the user device 100 and executes it on the processor unit 101, the application program 109 loads the shortcut operation program component 111 into the processor unit 101 for execution, and the application program 109 will use the shortcut operation program component 111 On the touch screen unit 107 Generate a shortcut interface, preferably such as but not limited to: a series of graphical user interface (GUI), quick menu (quick menu) based on one-click operation, quick operation webpage, quick list, Shortcut buttons, a graphical shortcut interface, a floating convenient option menu window, a drop-down option convenient menu, or a shortcut option menu are provided for the user 150 to click and operate through the touch screen unit 107 .

The front-end application program can be applied to Android and iOS systems, two major device platforms for smartphones currently on the market. For the Android device part, the front-end application program is developed using the Android Studio design editor, and the remote server program is compared with the software system part. It is best to use such as but not limited to PHP scripts to access the relational database MySQLDB to send and receive data. MySQLDB is a multi-user, multi-threaded SQL database server that can be used effectively for a database software. Arranging, documenting, and tabulating, so as to query, organize, send and receive data more efficiently in the future; in the part of iOS devices, the application program is preferably developed and designed using basic components such as but not limited to Xcode and UIKit (App ) interface, using Objective-C programming language or Swift programming language for development.

The processor unit 101 receives the instructions issued by the user through the front-end application program, and according to the instructions issued by the user, connects to the system server 200 through the radio frequency communication module 103, and accesses the sarcopenia database 250, including reading (read) required data or writing (write) data, and downloading to the user device 100 of the client, through the processing of the application program 109 and the shortcut operation program component 111 contained in the application program 109, and finally The sarcopenia database 250 can be built and stored in the internal storage unit of the system server 200, or it can be connected with the system server 200 and displayed to the user on the touch screen unit 107 through a graphical user interface or a shortcut menu. The device 200 is constructed separately and otherwise stored on other separate devices. Applied on user equipment 100 Between the program 109 and the sarcopenia database 250 of the system server 200, the communication protocol such as but not limited to HTTP/HTTPS is selected, and the data exchange and two-way communication are carried out in cooperation with the format such as but not limited to JSON.

The application program 109 and the shortcut operation program component 111 or other program components contained in the application program 109, as well as the remote server program or software system, can be programmed and implemented using any programming language, such as but not limited to C, C++, C#, Java, VBScript, Macromedia Cold Fusion, COBOL, Microsoft Dynamics Server Web pages, assembly languages, PERL, PHP, awk, Python, Visual Basic, SQL stored procedures, PL/SQL, any UNIX command description language, and a data structure, Extensible Markup Language (XML) for various algorithms implemented by any combination of objects, procedures, routines, or other program elements.

In one embodiment, the program components, modules, or software engines configured on the user equipment 100 can be implemented in the form of applications 109 or micro-apps (micro Apps), and these applications 109 or micro-apps can be implemented in Execute in the mobile operating system environment, mobile operating systems include, for example: Palm mobile operating system, Windows mobile operating system, Android operating system, Apple iOS, Blackberry operating system, etc.

The sarcopenia assessment system proposed by the present invention, in the part of the front-end program, in addition to being built as an application program for users to operate, it can also be deployed in an off-premises manner, and can also be implemented as a software-as-a-service (SaaS) or Platform-as-a-service (PaaS) cloud technology construction, and through the web browser (internet browser) executed on the user equipment 100 as the front-end user interface, it is directly provided to the user for operation, and provides the user with shift work Service, in the case of using SaaS or PaaS services, users only need to obtain permission to access and use the system online, and users do not need to install applications on mobile devices.

The sarcopenia assessment system of the present invention and the included application programs are composed of various hardware in the physical equipment layer and application programs/software platforms/computer program products in the application program layer, according to the international open system interaction Even with the communication reference model (OSI/RM) architecture, the application program of the medical and sanitary material shift operation of the present invention and the shortcut function program components contained therein are executed and operated on the seventh layer (application layer) of the OSI/RM architecture Software application service, the software application service at the 7th layer can independently select various communication protocols in the 4th layer transport layer, form a data packet and determine the transmission path at the 3rd layer network layer, and add it through the 2nd layer data connection layer After Logical Link Control (LLC) and Media Access Control (MAC), various devices located on the physical layer of the first layer, such as but not limited to multiple user equipment 100, smart phone 130, tablet device 135, The system server 200 and so on establish required uplink and downlink communication links (upload and download communication links).

The sarcopenia assessment model proposed by the present invention is preferably a machine learning classifier based on artificial intelligence, covering supervised classifiers and non-supervised classifiers As for the establishment, the establishment step at least includes items such as a sarcopenia action questionnaire survey procedure, a data preprocessing procedure, a labeling procedure, a training procedure, a verification procedure, and a testing procedure.

Firstly, implement the sarcopenia mobile questionnaire survey program, collect a large amount of raw data (raw data) as model training data, and provide mobile questionnaires through applications or web browsers as information and communication technologies such as smart phones and tablet computers deepen in life The practice of using tables can reach a large number of respondents very effectively, help to systematically and efficiently collect a large amount of required original data, and overcome the problem that large-scale questionnaire surveys are not easy to implement.

The present invention refers to the guidelines provided by the App Rating Scale (MARS) technology, and builds the application program and user interface (UI) required for the sarcopenia action questionnaire survey, and provides web pages The user interface loaded by the browser, etc., and the relevant sarcopenia action questionnaire was designed. After providing a large number of users to download and answer through mobile devices, it can effectively complete a large number of sarcopenia original sample surveys in a relatively short period of time. data collection work.

MARS technology divides applications into five quality levels, which are functionality, information quality, aesthetics, subjective quality, and engagement. The basic evaluation can evaluate mobile health applications that are suitable for patient disease awareness and nursing guidance care, and the results can also be used as a reference for application developers to design related applications in the future.

The content of the Sarcopenia Action Questionnaire is designed according to the 5-point scoring method proposed by the Likert Scale principle, in which strongly disagree is 1 point, disagree is 2 points, common is 3 points, Agree with 4 points and strongly agree with 5 points. The designed questionnaire was also handed over to a number of experts, such as medical information experts, engineering professors and information engineers, to verify and revise the appropriateness and clarity of the questionnaire content. To ensure the expert validity of the questionnaire, the action questionnaire of the present invention, its overall validity index (Total CVI) is controlled to at least reach 0.90, and the Cronbach's α value of internal consistency is greater than 0.9.

For example, the Sarcopenia Behavior Questionnaire designed according to the MARS technology and the Likert Scale principle of the present invention, its content is preferably as shown in Table 1 below:

The Sarcopenia Behavior Questionnaire designed by the present invention, in the total scoring method of MARS, is divided into two partitions for functionality (functionality) and subjective quality (subjective quality), and the cumulative total score is 0-20 points , the average total score for functionality is 19.16±1.838, and the average total score for subjective quality is 19.15±19.13. Through the use of mobile questionnaire technology, large-scale mobile questionnaire surveys can be efficiently carried out, and the difficulty of implementing large-scale questionnaire surveys in conventional technologies can be overcome.

After the Sarcopenia Action Questionnaire survey is completed, the collected raw data must be pre-processed first, including outlier deletion, error data deletion, incomplete record elimination, field Recognition, text recognition, semantic recognition, format conversion, standardization or formatting, etc., to remove incorrect and missing records, unreliable and possibly untrue Answers, etc., only keep correct and reliable records, and complete the pre-processing of the data. Since the present invention uses an application program or a web browser to provide mobile questionnaire surveys, the collected raw data have been stored and recorded in the cloud database in the form of digital files, and the traditional paper questionnaire survey can be omitted. In the process of converting this document into digital data, problems such as data errors, loss, and identification errors may be encountered.

The results of the sarcopenia survey will include the responses of many respondents. The more respondents the better, and it should preferably cover those who have been diagnosed with sarcopenia, those at high risk of sarcopenia, those at low risk of sarcopenia, and The absence of sarcopenia, the presence or absence of chronic diseases, the types of chronic diseases, and those taking drugs for a long time will help to expand the number of maternal samples and make the characteristics of the data included in the overall data approach unbiased. In this example, a total of 112 people were collected, among which the presence or absence of chronic diseases, types of chronic diseases, and long-term medications were collected. Among them, there were 10 types of chronic diseases and 10 types of medications for long-term use, which were respectively compared with sarcopenia screening. The three major items tested: calf circumference, muscle strength (hand grip strength) and mobility. The chi-square test of statistical analysis was used to analyze the correlation and risk estimation of sarcopenia. Grip strength) and the probability of occurrence of three low normal values of mobility have a significant impact.

The characteristics of the original data collected are as follows. The average part of the calf circumference, muscle strength (hand grip strength) and mobility of the interviewees are listed in Table 2 below. The average calf circumference of the interviewees was 34.88±4.62 cm, and 77.7% of them had lower legs than normal, accounting for the majority. A total of 25 people were below the normal value (boys less than 34 cm/girls less than 32 cm), accounting for 22.3% of the respondents. The average value of muscle strength (hand grip strength) was 23.88±10.25 kg, and most of them fell below the normal value, accounting for 68.8%. A total of 35 people accounted for 31.3% of those who were lower than the normal value (less than 26 kg for boys; less than 18 kg for girls). The average force was 1.01±0.23m/s, 18 people (16.1%) were less than the normal value, and the rest were within the normal range, accounting for 83.9%.

The risk estimates related to the calf circumference of the respondents and chronic diseases and long-term drug use are listed in Table 3 and Table 4 below. Among the 10 types of chronic diseases and long-term medications, diabetes has a significant impact on the occurrence of low-normal calf circumference (p=0.005, OR=6.378), which means that patients with diabetes have a lower-than-normal calf circumference than those without diabetes The value is 6.378 times more risky. Long-term use of hypoglycemic drugs also has a significant effect on the occurrence of low-normal calf circumference (p=0.025, OR=5.188), indicating that patients who take hypoglycemic drugs have a lower-than-normal calf circumference than those who do not take hypoglycemic drugs The value is 5.188 times more risky.

In the part of respondents' muscle strength (hand grip strength) and risk estimation of chronic diseases, it is listed in Table 5 below. Among the 10 chronic diseases, diabetes has a significant impact on the probability of low normal muscle strength (hand grip strength) (p=0.009, OR=5.407), which means that people with diabetes have higher muscle strength (hand grip strength) than people without diabetes. grip strength) was 5.407 times more likely to be below normal. And suffering from high blood pressure has a significant impact on the probability of low normal value of muscle strength (hand grip strength) (p=0.006, OR=3.375), which means that patients with hypertension have lower muscle strength (hand grip strength) than those without hypertension. The risk of being below normal was 3.375 times greater.

The part of the risk estimation related to the muscle strength (hand grip strength) of the respondents and long-term drug use is listed in Table 6 below. Among the 10 long-term drug types, long-term use of antihypertensive drugs also has a significant impact on the probability of low normal muscle strength (p=0.001, OR=4.500), indicating that patients who take antihypertensive drugs are more likely than those who do not take antihypertensive drugs , had a 4.5-fold greater risk of lower-than-normal calf circumference.

The parts of respondents’ mobility, chronic diseases and long-term drug-related risk estimates are listed in Table 7 and Table 8 below. Among the 10 chronic diseases and long-term drug types, suffering from diabetes Diabetes has a significant impact on the probability of low normal value of mobility (p=0.024, OR=4.780), which means that patients with diabetes have a 4.78 times higher risk of mobility than those without diabetes, and long-term use Hypoglycemic drugs also have a significant effect on the probability of low normal mobility (p=0.036, OR=5.086), which means that patients with diabetes have a 5.086 times higher risk of mobility than normal than those without diabetes .

Figure 3 reveals the distribution of data features revealed by the unsupervised machine learning classifier of the present invention to extract calf circumference lower than the normal value raw data; Figure 4 reveals the muscle strength (hand grip strength) extracted by the unsupervised machine learning classifier of the present invention The data feature distribution map revealed by the raw data below the normal value; Fig. 5 shows the data feature distribution map revealed by the unsupervised machine learning classifier of the present invention that the extraction power is lower than the normal value raw data.

In this embodiment, the original data is input into the machine learning unsupervised classifier, the features contained in the original data are extracted, and the calf circumference, muscle strength (hand grip strength) and mobility risk estimation of the interviewees are analyzed to establish the risk of sarcopenia The basic classification and prediction model of the evaluation; for example, the calf circumference is smaller than the normal value, the muscle strength (hand grip strength) is smaller than the normal value, and the mobility is smaller than the normal value as the target variables of the classifier, age, gender, BMI, height Blood pressure, glaucoma, liver disease, sciatica, diabetes, hyperlipidemia, asthma, rheumatoid arthritis, gastroesophageal reflux, cardiovascular disease, heart disease Vascular drugs, antihypertensive drugs, liver drugs, stomach drugs, hypoglycemic drugs, hormonal preparations, hypolipidemic drugs, sleeping pills, painkillers, and traditional Chinese medicine are set as independent variables of the classifier, and the overall accuracy of the model is lower than normal. The accuracy of the model is 73%, the accuracy of the muscle strength (hand grip) is 66.7% lower than the normal value model, and the accuracy of the mobility model is 75.7% lower than the normal value model. Then input the original data into the machine learning classifier to extract the feature distribution map of the data. The extracted feature maps are shown in Figures 3 to 5.

According to the disclosure in Figure 3, it is obvious that for the group whose calf circumference is lower than the normal value, the most important data characteristics are age and body mass index (BMI). For the group whose muscular strength (hand grip strength) is lower than the normal value, the most important data characteristics are also age and body mass index. According to the disclosure in Figure 5, it is clearly shown that for the group whose mobility is lower than the normal value The most important data characteristics were also age and body mass index.

Then implement the labeling procedure. The labeling procedure is to retain the correct data after data pre-processing. According to the pre-defined rules for the answer results of each interviewee, such as but not limited to: European Union Sarcopenia Working Group (EWGSOP) The operational definition of sarcopenia, the operational definition of sarcopenia of the International Working Group on Sarcopenia (IWGS), the operational definition of sarcopenia of the Asian Working Group on Sarcopenia (AWGS), the results of preliminary statistical analysis, the distribution map of data characteristics or use The operational definition of sarcopenia can be customized by the interviewee, etc., and each interviewee is given a different risk level of sarcopenia.

For example, after selecting the predefined rules to be followed, in accordance with the operational definition rules of EWGSOP sarcopenia, the risk of sarcopenia is further divided into two levels such as normal and high risk, or the risk of sarcopenia is divided into Three levels of low risk, medium risk, and high risk, or divide the risk of sarcopenia into more risk levels, and mark each respondent in the original data according to the operational definition of sarcopenia in the original data EWGSOP .

Tables 9 and 10 list the operational definitions and recommended screening methods for sarcopenia by EWGSOP, IWGS, and AWGS.

It is better to use statistical software (SPSS 24.0/G-power3.1) for statistical calculation of the above-mentioned statistical analysis, set the α value to 0.05, the detection power to 0.8, and the effect size to 0.3 to estimate and estimate the number of samples 100 people, considering the 20% attrition rate, the estimated total sample size is 120 people.

Finally, split the training dataset and the test dataset, where the training dataset contains the validation dataset, and the labeled original data text, The ratio of roughly 1:1 is divided into training set and test set, which are used to fine-tune the sarcopenia evaluation model. After the model learns and interprets the text of the marked answer results, the construction of the sarcopenia evaluation model can be completed. In this implementation The example system collects more than 1,000 answer results for training.

The machine learning classifier used in the present invention is preferably for example but not limited to: supervised classifier, unsupervised classifier, regression tree classifier, random forest (random forest) classifier, decision tree classifier, boosting tree (boost tree) classifier, gradient boosting tree classifier, strong gradient boosting machine classifier, weak gradient boosting machine classifier, ensemble learning (ensemble learning) classifier, or support vector machine (SVM), etc.

In summary, the sarcopenia assessment model proposed by the present invention is essentially a multi-class classification, and the supervised machine learning technology used in the sarcopenia assessment model is mainly based on the clustering method . Use the method of data grouping to distinguish the data, and establish similar clusters to carry out the grouping operation between the data, so that the data in the same cluster have the smallest difference and the highest similarity, and at the same time make the difference between different clusters the largest and the similarity Minimize, that is, use this technique to reduce the complexity of the data and find common features in the same cluster. The main purpose of grouping is to find out the differences between groups and the similarities in the same group, so that the differences within the group are small and the differences outside the group are large. At present, the most commonly used clustering analysis algorithms include two categories: segmented clustering algorithms and hierarchical clustering algorithms. The number of clusters in the present invention is known, and it is suitable to use the split clustering method. With such a system model, it can be used to estimate, predict, make decisions, diagnose sarcopenia, and evaluate the risk level for each user.

The sarcopenia assessment system of the present invention preferably applies SaaS and PaaS technologies, using the web browser executed on the user equipment 100 as a front-end user interface to provide the user with operations, or using an application program executed on the user equipment 100 Provided as a front-end user interface to use Operate by the operator, and generate a series of graphical user interfaces as shown in Figure 6 to Figure 8 through a browser or application program, and based on one-click operation with quick menu (quick menu) The operation shortcut interface is provided for users to operate and click, and provides users with sarcopenia self-assessment services.

Figure 6 discloses a schematic diagram of the sarcopenia self-test instruction interface displayed on the user equipment by the sarcopenia assessment system of the present invention; Figure 7 discloses the test-self-test displayed on the user equipment by the sarcopenia assessment system of the present invention Schematic diagram of the interface; Figure 8 discloses a schematic diagram of the sarcopenia assessment result interface displayed on the user equipment by the sarcopenia assessment system of the present invention; in this embodiment, the user equipment is preferably such as but not limited to a smart phone, And an application program installed on a mobile phone as an example.

On the touch screen of the smart phone 130, the user clicks on the sarcopenia application program, and clicks on the option to enter the self-test. First, a self-test instruction interface 310 as shown in FIG. 6 will be displayed. The self-inspection explanation interface 310 shown is a graphical user interface (GUI), and the system explains to the user the correct various sarcopenia detection methods through the content displayed in the self-inspection instruction interface 310, so that the user can operate and perform self-inspection . After the user understands the sarcopenia detection method, he can choose to perform the self-test of test 1 or test 2. The self-test instruction interface 310 provides a test 1 shortcut key 312 and a test 2 shortcut key 314. In this embodiment, the user chooses to execute Detect 1, and press the Detect 1 shortcut key 312 .

The system provides a test-self-test interface 320 as shown in FIG. 7 to assist the user to perform test 1. The test-self-test interface 320 provides multiple shortcut input fields, including at least the DAX value shortcut input field 321, and the calf area. The shortcut input field 323, the muscle mass shortcut input field 325, and the mobility ability shortcut input field 327 are for the user to follow the shortcut input field. Prompt, enter the information required by the sarcopenia assessment system and model, the system will continue to ask the user whether he has chronic diseases and whether he has taken drugs for a long time through other interfaces. After the user completes all input, press such as but not limited to Execute the detection shortcut key 329, the application program will send the input information back to the remote system server 200, and send a model execution command, requesting the sarcopenia assessment smart platform to execute the sarcopenia assessment model, and use it for evaluation by executing the sarcopenia assessment model Whether the patients belong to the high-risk group of sarcopenia.

The system server 200 returns the evaluation result to the smart phone 130 application program, and through the sarcopenia evaluation result interface 330 shown in Figure 8, the user is prompted with the evaluation result of the artificial intelligence model. According to the information input by the user, the disease assessment model judges that the user belongs to the high-risk group of sarcopenia, and displays such assessment results to the user on the sarcopenia assessment result interface 330, as well as relevant details including probability values and risk factors etc., for users to read and understand their own sarcopenia risk status.

Figure 9 discloses a flow chart of the implementation steps of the sarcopenia assessment method of the present invention; in summary, the sarcopenia assessment method 500 of the present invention preferably includes the following steps: implementing the sarcopenia action questionnaire survey program, through the front-end program to multiple A respondent provides a quick interface to conduct a sarcopenia action questionnaire survey, and obtain a sarcopenia survey result (step 501); input the sarcopenia survey result or the labeled data set into a machine learning classifier to train the machine Learning the classifier to establish a sarcopenia assessment model (step 503); providing users with sarcopenia autonomous detection through the front-end program, and obtaining the detection result (step 505); uploading the detection result to a machine learning classifier for construction The sarcopenia assessment model (step 507); and execute the sarcopenia assessment model at the remote end to implement sarcopenia assessment, so as to evaluate whether the user belongs to the sarcopenia high-risk group according to the detection result (step 509 ).

The above embodiments of the present invention can be arbitrarily combined or replaced with each other, thereby deriving more implementation forms, but none of them depart from the scope of protection intended by the present invention. More embodiments of the present invention are further provided as follows:

Embodiment 1: A sarcopenia evaluation method, which includes: providing users with sarcopenia autonomous detection through a front-end program, and obtaining detection results; uploading the detection results to a sarcopenia built based on a machine learning classifier an assessment model; and remotely execute the sarcopenia assessment model to implement sarcopenia assessment, so as to assess whether the user belongs to a high-risk group for sarcopenia according to the detection result.

Embodiment 2: The sarcopenia assessment method as described in Example 1, wherein the sarcopenia assessment model is established according to the sarcopenia assessment model establishment method, and the sarcopenia assessment model establishment method includes: implementing sarcopenia The mobile questionnaire survey program provides multiple respondents with a quick interface through the front-end program to conduct the sarcopenia mobile questionnaire survey, and obtains the survey results of sarcopenia; selectively implements the labeling program, and conducts the sarcopenia survey according to the predefined rules Annotate the results, classify the sarcopenia survey results into at least two types of results, and obtain a labeled data set; and input the sarcopenia survey results or the labeled data set into a machine learning classifier to train the machine learning classifier Establish a sarcopenia assessment model.

Embodiment 3: The establishment method of sarcopenia evaluation model as described in embodiment 2, this predefined rule is selected from European Union sarcopenia working group (EWGSOP) sarcopenia operational definition, international sarcopenia working group (IWGS ) sarcopenia operational definition, Asian sarcopenia working group (AWGS) sarcopenia operational definition and user-defined sarcopenia operational definition.

Embodiment 4: The sarcopenia evaluation model establishment method as described in embodiment 2, this machine learning classifier is selected from supervised classifier, unsupervised classifier, regression tree classifier, random One of machine forest classifier, decision tree classifier, boosted tree classifier, gradient boosted tree classifier, strong gradient boosting machine classifier, weak gradient boosting machine classifier, ensemble learning classifier, support vector machine and combinations thereof.

Embodiment 5: A sarcopenia assessment system, which includes: a system server, which is installed with a sarcopenia assessment smart platform including a sarcopenia assessment model built based on a machine learning classifier; and a user device, which It is configured separately from the server of the system and connected by communication, and the front-end program of the smart platform for sarcopenia evaluation is installed, so as to implement autonomous detection of sarcopenia to the user through a shortcut interface, and obtain the detection results, wherein the user's device The detection result is uploaded and input into the sarcopenia evaluation model, so as to automatically evaluate whether the user belongs to a high-risk group of sarcopenia according to the detection result.

Embodiment 6: The sarcopenia assessment system as described in Embodiment 5, the user equipment is one of a mobile device, a smart phone, a tablet device, a desktop computer, and a notebook computer.

Embodiment 7: The sarcopenia assessment system as described in Embodiment 5, the front-end program is one of an application program and a web browser.

Embodiment 8: The sarcopenia assessment system as described in Embodiment 5, the shortcut interface is selected from the graphical user interface, shortcut menu, shortcut operation webpage, shortcut list, shortcut keys, graphical shortcut interface, and floating convenient options One of a menu window, a drop-down option shortcut menu, a shortcut menu interface, a shortcut option interface, a shortcut option menu, and a shortcut selection button.

Embodiment 9: A method for establishing a sarcopenia assessment model, which includes: implementing the sarcopenia action questionnaire survey program, providing a quick interface to multiple respondents through the front-end program to conduct the sarcopenia action questionnaire survey, and obtaining the sarcopenia action questionnaire Survey results: Selectively implement the labeling procedure, mark the survey results of sarcopenia according to predefined rules, and distinguish the survey results of sarcopenia At least two types of results are obtained, and a labeled data set is obtained; and the sarcopenia investigation result or the labeled data set is input into a machine learning classifier, so as to train the machine learning classifier to establish a sarcopenia assessment model.

The various embodiments of the present invention can be combined or replaced arbitrarily with each other, thereby deriving more implementation forms, but none of them depart from the intended protection scope of the present invention, and the definition of the protection scope of the present invention is fully defined by the patent scope of the present invention application The recorder shall prevail.

500: sarcopenia assessment method of the present invention

501-509: Implementation steps

Claims (9)

A method for assessing sarcopenia, comprising: Provide a user with an autonomous detection of sarcopenia through a front-end program, and obtain a detection result; uploading the test result to a sarcopenia assessment model built based on a machine learning classifier; and Executing the sarcopenia assessment model remotely to implement a sarcopenia assessment, so as to assess whether the user belongs to a high-risk group of sarcopenia according to the detection result. The sarcopenia assessment method as described in Claim 1, wherein the sarcopenia assessment model is established according to a sarcopenia assessment model establishment method, and the sarcopenia assessment model establishment method includes: Implement a sarcopenia action questionnaire survey procedure, provide a quick interface to multiple respondents through a front-end program to conduct a sarcopenia action questionnaire survey, and obtain a sarcopenia survey result; Selectively implementing a labeling procedure, labeling the sarcopenia survey results according to a predefined rule, classifying the sarcopenia survey results into at least two types of results, and obtaining a set of labeled data; and Inputting the sarcopenia investigation result or the labeled data set into a machine learning classifier to train the machine learning classifier to establish a sarcopenia assessment model. The method for establishing a sarcopenia assessment model as described in claim 2, wherein the predefined rule is selected from an operational definition of sarcopenia of the EU sarcopenia working group (EWGSOP), an international sarcopenia working group (IWGS) Operational Definition of Sarcopenia, an Asian Working Group on Sarcopenia (AWGS) Sarcopenia One of an operant definition and a user-defined sarcopenia operant definition. The method for establishing a sarcopenia assessment model as described in claim 2, wherein the machine learning classifier is selected from a supervised classifier, an unsupervised classifier, a regression tree classifier, a random forest classifier, a One of a decision tree classifier, a boosted tree classifier, a gradient boosted tree classifier, a strong gradient boosting machine classifier, a weak gradient boosting machine classifier, an ensemble learning classifier, a support vector machine and combinations thereof . A sarcopenia assessment system comprising: A system server installed with a smart platform for sarcopenia assessment including a sarcopenia assessment model built based on machine learning classifiers; and A user device, which is separately configured and communicated with the system server, and is installed with a front-end program of the sarcopenia assessment smart platform, so as to implement a sarcopenia autonomous detection for a user through a shortcut interface, And get a test result, Wherein the user device uploads the detection result and inputs it into the sarcopenia evaluation model, so as to automatically evaluate whether the user belongs to a high-risk group of sarcopenia according to the detection result. The sarcopenia assessment system as described in Claim 5, wherein the user equipment is one of a mobile device, a smart phone, a tablet device, a desktop computer, and a notebook computer. The sarcopenia assessment system as described in Claim 5, wherein the front-end program is one of an application program and a web browser. The sarcopenia assessment system as described in claim 5, wherein the shortcut interface is selected from a graphical user interface interface, a shortcut menu, a shortcut operation page, a shortcut list, a shortcut button, a graphical shortcut interface, a floating convenient option menu window, a pull-down option shortcut menu, a shortcut menu interface, a shortcut option interface, One of a shortcut option menu and a shortcut selection button. A method for establishing a sarcopenia assessment model, comprising: Implement a sarcopenia action questionnaire survey procedure, provide a quick interface to multiple respondents through a front-end program to conduct a sarcopenia action questionnaire survey, and obtain a sarcopenia survey result; Selectively implementing a labeling procedure, labeling the sarcopenia survey results according to a predefined rule, classifying the sarcopenia survey results into at least two types of results, and obtaining a set of labeled data; and Inputting the sarcopenia investigation result or the labeled data set into a machine learning classifier to train the machine learning classifier to establish a sarcopenia assessment model.

Images