TW201234309A - Physical fitness analyzer, analytic method and bicycled fitness device - Google Patents

Abstract

A physical fitness analyzing method is disclosed. The process of the physical fitness analyzing method includes some steps as following: Several physiological characteristic data are detected from a user's physiological status. Several electromechanical characteristic data are detected from an exerciser. At least one element model is generated according to the physiological characteristic data and the electromechanical characteristic data. The element models are applied to train an extension neural network (ENN) to establish diagnosis system. The ENN system is applied to determine diagnosis class of the exerciser.

Description

201234309 VI. Description of the Invention: [Technical Field of the Invention] The present disclosure relates to a method for assessing and analyzing physical fitness, and in particular to a method for assessing physical fitness in a foot exerciser. 】 Physical Fitness is defined as the ability of the individual to be able to enjoy leisure and to be able to cope with sudden changes and stresses. In addition to being able to adapt to the external environment, the physical fitness is defined as the physical ability. The comprehensive ability of factors such as temperature, climate change, virus, etc. In a civilized society with advanced science and technology, the opportunities for human physical activity are becoming less and less, the nutritional intake is getting higher and higher, and the pressure of work and life is relatively increased. Therefore, proper exercise is the main way to improve the health of modern people. People with better physical fitness have better vitality and adaptability in their daily activities or work, and they do not easily feel tired or unable to feel at ease. In order to achieve a healthy life, modern people often have the habit of using exercise equipment to exercise. It is known that the pedal exerciser can initially measure the user's heartbeat and present it on the screen; however, this function can only be used faithfully. At present, the state of using fitness equipment cannot be further evaluated and monitored for individual users. [Summary of the Invention] 201234309 Evaluation and Analysis Party ί:: The second is to provide - the evaluation and monitoring of the physical fitness users, = for each pedal exerciser according to the technical aspect of the implementation method , poetry at least - make ^ W _ fitness assessment step by step: when recruiting multiple physiological characteristics, including the physiological state of the user, take a number of electromechanical; = special = material from the data collected from the pedal fitness The electromechanical = special ==, the description of the electromechanical special electromechanical characteristics data to establish at least one matter:: the characteristics of the information and the category.捋 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 知 至少 至少 至少 至少 至少 至少 至少 至少 至少 至少 至少 至少 至少 至少 其他 其他Output voltage characteristics and "force = or: horse utilization - remote monitor collection evaluation _. 彳 surface, its analyzer can be used for two - body fitness assessment _ this technical aspect - real::; = = analysis The method comprises: a detecting unit, and a seeding and fitness evaluation and separating unit. The measuring unit is configured to acquire a plurality of raw systems and an input number to establish a data to be tested. Xi Li Xinhu and a number of electromechanical fitness fitness assessment methods are trained, using the second > and - evaluation categories. Finally 'the round-out unit is used to output evaluation == 201234309 This technical aspect in other implementations The fitness assessment analyzer can further be a remote monitor to collect the aforementioned evaluation categories. The technical aspect of the present disclosure is to provide a fitness assessment function for the user to operate the pedal exerciser. According to an embodiment of the present invention, an exercise device for pedaling is proposed. Include a foot pedal = ^ can evaluate the detector, multiple electromechanical signals _ device and a plurality of physiological signals. The physiological signal detector is used to capture the foot = = : estimated the state of the division, electrical and mechanical signal detection ^^ The user's ecology is stored, and the fitness assessment analyzer is used to construct the electromechanical signal detection H of the body, and to evaluate the physical assessment device and the electromechanical at least one evaluation category. ° The data to be tested is judged to establish an extensional neural processing system, and the extensional neural network: the neural processing system for each fitness equipment: that is, the remote monitoring device can collect the pings of each In addition, it does not 'long-term monitoring' to achieve long-term suspension: = the purpose of the class T [implementation] method of physical fitness assessment method analysis method, at least include the physical fitness of the lower application method Ding _ a physiological characteristic data, the physiological state of the aforementioned physiological characteristics 201234309. Further, if the sacral data is as described above, the electromechanical state of the electromechanical feature is as follows, as shown in step 130, the electromechanical state of the pedaling exerciser Material establishment at least - matter element model Feature data and electromechanical characteristics of the material element model training one extension:: [4], using the system. Finally, if the steps are taken to establish a known neurological treatment method, the user of the fitness equipment can be smeared in advance. Evaluation:: Remuneration, in order to quickly identify the cautionary ' 'Step 110 to measure the user's various physiological characteristics steps, m characteristics, EMG signal characteristics, etc.; on the other hand, stepping _ Γ: Ma::=种=_;", such as feet in this implementation, temple, etc. As for the measurement method, there are many options, 41⁄23⁄4 _ 疋 疋 疋 生理 生理 生理 生理 生理 生理 生理 生理 生理 生理 生理 生理 生理 生理 生理忾 较 较 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷 陶瓷Standing at least one: 4: GO uses the physiological characteristics data and electromechanical characteristics data to build the jump rate ^ production of the model, the matter model in this embodiment has the heart to step on the number of dehydration, knee muscle EMG, conversion power and Efficiency and class gods ^ network ^ down 'step 140 using the matter-element model to train a extension a 4|, a dual-extension-type neuroprocessing system. Finally, 'Step 150 Dan's use of extension class to draw less-evaluation _ 4 system to classify at least one data to be tested as to know 'together*' The evaluation category of the data to be tested can be divided into heart ages of 201234309 degrees, Replenish water tips, knee muscle fatigue, total calories burned energy, and riding techniques. Its towel, heart age production can be subdivided into basic physical strength, weight loss body sculpting, cardiopulmonary function, aerobic boundary 2 and competitive physical fitness; riding technology can be subdivided into comfort, exercise, competition: and winning conditions. Λ Μ In order to calculate the data to be tested and the physical fitness assessment method of the present embodiment based on each physiological characteristic data and the electromechanical characteristic data, the classification of the matter element model, that is, the data to be tested should be classified. For which type of call evaluation category, this embodiment refers to the extension theory for the evaluation of the standby. The specific operation principle is explained as follows: Figure 2 shows the architecture diagram of the extension-like neural network. The extension-like neural network has the ability to accept different kinds of denaturations, which include the input layer, the calculation layer, and the number of rounds to be the first to lose; = t input and classify the constituent element module into the person (four), the number of input layers The characteristics of the matter-element model are calculated from the data and stored in the second:, &, * output layer belongs to the output layer of each category to calculate the distance value that can be included in the evaluation category of the measured data. The study method of the extension-type neural network can be divided into A-style learning (Surveyed, non-supervised learning and supervision of the characteristic sample values to learn, by; currently has relevance, when there is - The information needs to be identified by the person to find the regularity of the skill and the result of the identification. The present embodiment uses:: find the most: the supervisory learning, the supervised learning is through the network ^ use the ground learning And training to adjust the correction weight == this::: 201234309 = the difference between the output value of the neural network and the target output value, thus the accuracy of the neural network identification. Therefore, you must have a S sample before learning. x={Xl,x2,X3”..Xpm}, material and category (4) In II =: represents' samples compared to the total number of PM, the total number of the total difference was 4PN wherein m, the total error rate is set Ετ, as shown in Formula ⑴..

ET

The calculation procedure of Pm (1) and extension-like neural supervised learning is as follows: Then Si 1: The learning material is used to establish the weight values of the input and output using the matter-element model, and the representation of the matter element is as follows: N/ c,c” Vk' ci,Κι •♦· ··· Cn,Κη_ (2)

The number of categories in which the special (^) towel represents the data, Gj is the nth η in the matter element model, and 』1'2'3'...'η,, =<w/,w/> The classic domain of & and the classical domain range can be determined by the learning data as follows: J κ = k) N (3) (4) where 'Xijk stands for the input data of the extension-type neural network. Step 2: Calculate each The weight center value of the item feature is as follows: ', (5) (<+<) 2 (6) 9 201234309 Where, if the learning material has only the same set of data in the same feature, because the upper limit of weight will be Since the lower weight limits are equal, the equations (7) and (8) are used to adjust to avoid infinitely large values of the extension distance. The adjustment formula is as follows:

Wkj_new = Wkj_old - (〇r X 2 Wkj_new = wkj_〇id + (ax ζ; (7) (8) , α is the classical domain range adjustment rate. When the α setting is larger, the classical domain range is also increased. 3: Read the i_th training sample data and the feature number k as follows: (9) ED)^ Use ^ to calculate the extension distance one by one, EDik ~·|·

Vt . wk) (10) Figure 3 shows the extension distance diagram. As shown in Figure 3, the distance between the extension and the Fan® <"^" is greater when the eight-domain range of the eigenvalues is larger, and the calculation of the extension is 3: The lower; the opposite 'if the classical domain range of eigenvalues is smaller, the more accurate the data sample is, and the higher the sensitivity. = 5·· Find the category of the most * expandable distance of all categories from the minimum extension distance (4) Break into the second data category, ..., then skip to step (11) 201234309 Equal k* off k ' Then continue to step 6 The action. Step 6: Adjust the weight values of the k category and the k* category. (1) Update the weight upper and lower limit values. (12) (13) (14) 〇 5) (16)

Wkj_new ~ Wkj_old +7l(Xij ~ Zkj_old) Wkj_new ~ Wkj_old + ~ Zkj_〇ld)

Aj= Aj_M «厂'·』

WH(4-zk"_old, (2) update the weight center value. =^kj_new ^kj^new) Δ kj new ; · 2 where 77 is the learning rate, and the learning rate affects the convergence rate. And the accuracy of convergence, if the learning rate is larger, it is easy to reach convergence, but the accuracy of convergence may also be reduced. Conversely, if the learning rate is smaller, the convergence is more accurate, but the number of learning and time may increase. Figure 4 is a schematic diagram showing the k-category and k*-category weight adjustment, and Figure 5 is a schematic diagram showing the k-category and k*-category weight adjustment. As shown in Figures 4 and 5, 'EDk*_old < EDk_old, which means that the category judged is not the category of the data'. After adjusting by equations (10) to (16), the EDk*_new > EDk_ new of the learning data in Fig. 5 indicates By adjusting the weights, the categories to which they belong have been changed to the correct category. Step 7: Repeat steps 3 through 7 until all learning 201234309 data is read and completed. Step 8: When all is completed. Total error rate is reached. The target value has reached the convergence state. After the above steps, go back to step 3. Continue: Sub-acquisition and training under the 'corrected weight: middle learning and adjustment weight value, from the output value and target C rate' can reduce the extension type neural network identification The accuracy rate. The gap between the two can be improved. The green picture of Figure 6 can be improved. The understanding of ST can be: learning flow chart. For example, _ two to %, in this embodiment, the way of learning and how to The circle 'and the identification rate of the liftable appreciating evaluation category. When the extension type neural network is finished, the human body physiological signal identification is evaluated. = The identification of the set category and: the round-up identification process of the data to be tested. As shown in Figure 7 and Figure 7, the neural network can be read as follows: The reading process is as follows: The multi-value matrix. The purpose of the extension-type neural network. Step 2: Calculate the intermediate value. Step 3: Read the data to be tested Step 4: Calculate the test to be tested Step 5: Find the minimum distance that can be used to estimate the category. The evaluation category. Moment away, to judge the data to be tested belongs to step 6 · Identify all the defects Step 3: Read the next data to be tested. Calculate 'otherwise, the above is the operating principle and mode of the present embodiment. It is worth = the specific method of evaluating the analysis method 100, and the unique neuron system can be established by the extensional neuroprocessing system 201234309 system, and accurately Each user's unique physical performance, and then to assess its most accurate physical fitness. Figure 8 shows the functional block diagram of the fitness assessment analyzer. As shown, from a technical point of view, The fitness assessment method 100 can also be embodied as a fitness assessment analyzer 200, which includes a detection unit 210, an extension-type neural processing system 220, and an output unit 230. The detecting unit 210 is configured to capture a plurality of physiological signals 211 and a plurality of electromechanical φ signals 212 to establish a data to be tested. The detecting unit 210 here corresponds to the aforementioned physiological detector, ceramic sensor, electrode patch, and the like. Among them, the present embodiment uses some biomedical techniques to measure some physiological signals obtained by the human body during exercise, and uses these physiological signals 211 to find physiological characteristic data. In order to find out these physiological characteristic data, the following respectively No. and EMG signal introduction. Figure 9 shows the flow chart of the ECG signal. As shown in Fig. 9, the heart is an organ composed of muscles. The muscles that make up the heart are called myocardium. φ When the myocardium is active, that is, when the heart beats, current is generated, and in the static state, a ceramic sense is used first. The measuring device measures the user's heart current, and then measures the user's heart current at any time during the movement, after a series of lead selection circuits, high and low pass filter amplifiers, isolation circuits and filters, and finally obtains a single heart telegram. number. Then use these ECG signals to determine the age of the heart is basic physical strength, weight loss body sculpting, cardiopulmonary function, aerobic boundaries, competitive physical fitness and so on. Figure 10 shows a flow chart of the acquisition of myoelectric signals. As shown in Figure 10, when the human body is exerting force, the cells in the muscle are excited by the stimulation. 13201234309=The motion is contracted, and the surface electrode is used to measure the high-low filtration of the myoelectric signal and the whole two-string. Wave Ail, _ circuit, filter 々 'final can get a muscle force diagram. Where = will, 1 electrogram. Through the muscle force diagram == strong production to t off the potential activity of meat removal, and by (4) for time, shape ^ 77 kg muscle is good or fatigue. , continue to refer to Figure 8, the fitness evaluation analyzer fine extension class

L i /1 μ 220 is trained in the physical fitness assessment analysis method 100 as described above to classify the data to be tested into at least one evaluation category. = unit 230 is used to output the evaluation category, where the output unit 23: two, type computer, pDA, mobile phone, USB ... any data output, I side - in addition, the embodiment further includes a remote monitor 240 For collection, the different evaluation categories of users and users are transmitted to the remote monitoring If 24G for control by Modbus and Zigbee wireless transmitter. The 11th to 13th drawings show the interface of the graphical control software LabVIEW. As shown in Fig. 11 and Fig. 13, this embodiment uses a notebook computer as a data storage and data output interface, so that the monitor can easily know the user's parameter values and evaluation from the LabVIEW interface. Category results. In this way, the user can also know his or her physical condition and the data converted during the exercise. The embodiment has the design of the remote monitor 24〇, and can be applied to the patient who is still doing long-term rehabilitation, and the remote monitoring unit 24 uses the evaluation category of the long-term collection to utilize the data statistics and graph analysis. The physical condition of the healthy patient and the rehabilitation process are closely monitored 'avoiding the rehabilitation of the sports injury of 201234309' and ensuring that the rehabilitation exercise has the desired effect on the patient's body. In addition, the present embodiment can be connected to a plurality of output units 23() by signal, and then manage multiple sets of evaluation categories of different users. This unified management method can be extended to various fitness centers, hospitals or rehabilitation centers, not only in fitness centers, hospitals or rehabilitation centers. It can improve the quality of the user's fitness or rehabilitation, and can also reduce the personnel costs of monitoring the user. Furthermore, the above-mentioned fitness evaluation analyzer can be applied to the pedal exerciser to make it a fitness evaluation and analysis pedal exercise device 300. Figure 14 is a schematic diagram showing the physical fitness assessment and analysis of the pedal exercise device. As shown in Fig. 14, the physical fitness evaluation analysis pedal exercise device 300 includes a pedal exercise device 310, a plurality of physiological signal detectors 32, a plurality of electromechanical signal detectors 330, and a physical body as described above. The analyzer 200 can be evaluated. " Physiological signal detector 320 is used to capture the use of the pedal exerciser 31

The physiological state of the person. The electromechanical signal detector 33 is used to capture the electromechanical state of the pedal exerciser 310. Here, the physiological signal detector 32 and the electromechanical signal detector 330 are equivalent to the detecting unit 2 of the aforementioned fitness evaluation analyzer 2). The at least one evaluation category is determined according to the physiological signal detector 320 and the electromechanical signal detector 330, as described above. Worth-k, the physical fitness evaluation of the present embodiment analyzes the pedal exercise device 300. Another feedback-driven generator system, the force of the pedaling wheel drives the chain and the transmission to start the generator, and the generator is electrically The battery can be stored in the battery by using a DC charger and a frequency converter, and then the battery is f (four) to the physiological signal _ 32G, the electromechanical signal detector 201234309 330 and the fitness evaluation analyzer 2 (eight). The second disclosure of the device can be used to collect the user's other = and = ' for each health device and to monitor the long-term monitoring by the remote monitoring of the physical fitness training or rehabilitation. The above is disclosed in the above embodiments, but it is not intended to show that the content of the present invention can be changed and retouched without departing from the disclosure of the present disclosure. ', _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Architecture diagram.

Figure 3 shows a schematic diagram of the extension distance. ^ 4 is a schematic diagram showing the k-category and k*-category weight adjustments. Figure 5 is a schematic diagram showing the weight adjustment of the k category and the k* category. Figure 6 shows the flow chart of the learning of the extension class. ^ 7 Figure drawing extension chart _ road identification flow chart. Figure 8 is a functional block diagram of the Green Inappropriate Assessment Analyzer. Figure 9 shows the flow chart of the telecentric system. The first diagram shows the flow chart of the myoelectric signal. ^ 11 to 13 show the interface of the graphical software LabVIEW. Figure 14 shows the light assessment analysis (4) Schematic diagram of the fitness device 〇201234309 [Main component symbol description] 100: Physical fitness assessment analysis method 110-150: Step 200: Physical fitness assessment analyzer 210: Detection unit 211: Physiological signal 212 ·•Electromechanical signal 220: Extension-type neuroprocessing system 230: Output unit 240: Remote monitoring device 300: Physical fitness assessment analysis Foot pedal 310: Foot fitness device 330: Electromechanical signal detector 320: Physiological signal detector 17

Claims (1)

201234309 VII. Patent application scope: 1. A physical fitness assessment analysis method for at least one user to operate a pedal exerciser, comprising the following steps: capturing a plurality of physiological characteristic data, the physiological characteristic data system Taking the physiological state of the user; taking a plurality of electromechanical characteristics data taken from the electromechanical state of the pedal exerciser; using the physiological characteristic data and the electromechanical characteristic data to establish at least one a matter-element model; using the matter-element model to train an extension-like neural network to establish an extension-type neural processing system; and using the extension-type neuroprocessing system to classify at least one data to be tested into at least one evaluation category . 2. The method of assessing physical fitness according to claim 1, wherein the physiological characteristic data is an electrocardiographic signal feature. 3. The method of assessing physical fitness according to claim 1, wherein the physiological characteristic data is a myoelectric signal feature. 4. The physical fitness assessment analysis method of claim 1, wherein the electromechanical characteristic data is a one-step tread speed characteristic. 5. The physical fitness assessment analysis method of claim 1, wherein the 201234309 electromechanical characteristic data is a motor output power characteristic. 6. The physical fitness evaluation method described in claim 1 is characterized in that the motor-electrical characteristic data is a motor output voltage characteristic and a horse ": levy 0, ❿ 7. The physical fitness as described in the claim 丨The evaluation analysis method further comprises: collecting the evaluation category by using a remote monitor. 8· A fitness assessment analyzer, comprising: a detection unit for extracting a plurality of signals to establish a (four) Lai; a physiological A The plurality of electromechanical systems are:; the t-type neuroprocessing system is based on the body-fitted b# estimation method as described in claim 1-7, and the less-evaluation category is formed; and the data to be tested is classified as a to-output unit. 'To output the evaluation category. The physical fitness evaluation analyzer described in claim 8 further includes: a monitor for collecting the evaluation category. The device can be evaluated, including: a plurality of physiological signal detector users Physiological state; used to capture the pedal exerciser 201234309 a plurality of electromechanical signal detectors for capturing the electromechanical state of the pedal exerciser; and the fitness assessment analysis as described in claim 8 Use Determining at least one category based on the evaluation of these physiological signal detector, and the plurality of electromechanical signal detector. 20 ΰ