Method for assisting diagnosis of osteoporosis by using plantar pressure

Abstract

In an aspect, the present disclosure relates to a method for assisting diagnosis of osteoporosis, comprising the steps of: measuring plantar pressure when a subject walks; and predicting the possibility of the subject being diagnosed with osteoporosis on the basis of the measured plantar pressure. In addition, in another aspect, the present disclosure relates to a computer program stored in a medium to execute the method for assisting diagnosis of osteoporosis, in combination with hardware. In another aspect, the present disclosure relates to an apparatus for assisting diagnosis of osteoporosis, performing the method for assisting diagnosis of osteoporosis, the apparatus comprising: a measurement module for measuring plantar pressure when a subject walks; and a prediction module for predicting the possibility of the subject being diagnosed with osteoporosis on the basis of the measured plantar pressure.

Description

Auxiliary method for diagnosing osteoporosis using plantar pressure

[0001] [Cross-reference to related applications]

[0002] This application claims priority to Republic of Korea Patent Application No. 10-2024-0018868, filed February 7, 2024, and Republic of Korea Patent Application No. 10-2024-0075951, filed June 11, 2024, the entire contents of which are incorporated herein by reference.

[0003] The present specification relates to an osteoporosis diagnosis assistance method using plantar pressure, a computer program stored in a medium for executing the osteoporosis diagnosis assistance method, and an osteoporosis diagnosis assistance device for executing the same.

[0004]

[0005] Osteoporosis is a skeletal disease in which bone mass decreases and bone strength weakens, making the bones prone to fractures. Osteoporosis is not a symptom itself, but rather a problem caused by various fractures, especially femur fractures or spinal fractures, that limit long-term activities due to bone weakening, and it is known to account for about 15% of deaths in the elderly.

[0006] Bone mineral density (BMD) accounts for approximately 70% of bone strength and is often used as a surrogate measure of bone strength and a predictor of fracture risk. Bone tissue responds to dynamic loading, whereas static loading does not initiate bone formation. Therefore, weight-bearing activities of daily living, such as walking, play a crucial role in maintaining bone density in ambulatory individuals. Because walking is the most common form of leisure activity, particularly among older adults, regular walking exercise is an important way to maintain bone density in older adults.

[0007] Many previous studies have reported a relationship between bone mineral density and various types of exercise, such as impact load exercise (jumping, weight-bearing exercise), and resistance exercise. These studies provide some insight into how the load applied to the body can affect bone mineral density.

[0008] Dual-energy X-ray absorptiometry (DEXA) is the gold standard for quantitative bone mineral density (BMD) measurement. However, it requires expensive heavy equipment, specialized operators, and radiation exposure, making it only available in hospitals equipped with the equipment. Given that osteoporosis affects up to 30% of postmenopausal women and that symptoms typically do not appear until a fracture occurs, screening or diagnosis for osteoporosis may not be appropriate.

[0009] Recent studies have shown that bone mineral density (BMD) is significantly correlated with gait data during comfortable walking. This suggests that BMD can be assessed using routine biometric data readily available during daily activities. However, 3D gait analysis still requires trained personnel with expensive equipment, including a charge-coupled device camera, force plates, and complex operating systems, to obtain gait data. This can hinder BMD screening during routine walking.

[0010] Healthcare broadly refers to comprehensive health management services that combine traditional treatment-based medical services with disease prevention and management concepts. With the recent rise in interest in health, a variety of healthcare-related devices, applications, and technologies are being developed to reflect the diverse needs of consumers.

[0011] Meanwhile, fueled by the rapid advancements in smart devices, technologies that enable connectivity between smart devices and other devices are being developed. For example, various smart technologies based on the Internet of Things are being developed, such as using smart devices to check the status of a vehicle or remotely control home appliances.

[0012] Considering this background, the inventors of the present invention analyzed the relationship between plantar pressure and osteoporosis and studied a method for predicting osteoporosis using plantar pressure during walking, thereby completing the present invention.

[0013]

[0014] In one aspect, the present invention seeks to provide a method for assisting in the diagnosis of osteoporosis using plantar pressure.

[0015] In another aspect, the present invention provides a computer program stored in a medium for executing a method for assisting the diagnosis of osteoporosis in combination with hardware, and an osteoporosis diagnosis assistance device for performing the same.

[0016]

[0017] In one aspect, the present invention provides an osteoporosis diagnosis assistance method, comprising: a step of measuring plantar pressure of a subject while walking; and a step of predicting the possibility of being diagnosed with osteoporosis based on the measured plantar pressure.

[0018] In another aspect, the present invention provides a computer program stored in a medium to execute an osteoporosis diagnosis assistance method according to one embodiment of the present invention in combination with hardware.

[0019] In another aspect, the present invention provides an osteoporosis diagnosis assistance device that executes a method according to one embodiment of the present invention, the device including: a measurement module that measures plantar pressure when a subject walks; and a prediction module that predicts the possibility of being diagnosed with osteoporosis based on the measured plantar pressure.

[0020]

[0021] In one aspect, the osteoporosis diagnosis assistance method according to one embodiment of the present invention can predict the possibility of being diagnosed with osteoporosis using plantar pressure.

[0022] In addition, the osteoporosis diagnosis assistance method and device of the present invention can help to quickly and easily predict the risk of osteoporosis diagnosis during daily activities.

[0023]

[0024] Figure 1 is a photograph showing the definition of the regions of interest of the lumbar spine, femoral neck, and total proximal femur.

[0025] Figure 2a is an image that distinguishes and displays eight sections of the sole of the foot used in the dynamic foot pressure test, and Figure 2b is a dynamic plantar pressure-time graph generated during walking according to the corresponding sections.

[0026] Figure 3a shows an ROC curve based on the maximum pressure of the 2nd and 3rd metatarsal heads for predicting osteoporosis, and Figure 3b shows a 2X2 confusion matrix table thereof.

[0027] Figure 4 shows the ROC curve combined with the weights of the ROC analysis.

[0028]

[0029] Hereinafter, exemplary implementation examples of the present invention will be described in detail.

[0030]

[0031] The terms used in this specification have been selected from widely used, current terms, taking into account their functions within the present invention. However, these terms may vary depending on the intentions of those skilled in the art, precedents, or the emergence of new technologies. Furthermore, in certain cases, the applicant may arbitrarily select terms, in which case their meanings will be described in detail in the relevant description of the invention. Therefore, the terms used in this specification should not be defined simply as names, but rather based on their inherent meanings and the overall content of the present invention.

[0032]

[0033] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Commonly understood terms should be interpreted as having the same meaning within the context of the relevant technology, and unless explicitly defined herein, they shall not be construed in an idealized or overly formal sense.

[0034]

[0035] In this specification, terms such as "module," "unit," "system," and "device" may refer to a combination of hardware and software driven by the hardware. For example, the hardware may be a data processing device including a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), or another processor. Furthermore, the software may refer to a running process, object, executable, thread of execution, program, etc.

[0036]

[0037] Each step or combination of steps in this specification may be implemented at least partially as a computer program and recorded on a computer-readable recording medium. For example, the program product may be implemented with a computer-usable or computer-readable medium containing program code, which may be executed by a processor to perform any or all of the described steps, operations, or processes. Furthermore, each step or combination of steps may be performed by computer program instructions (execution engine), and these computer program instructions may be installed in a processor of a general-purpose computer, special-purpose computer, or other programmable data processing equipment, so that the instructions executed by the processor of the computer or other programmable data processing equipment may create a means for performing the functions described in each step.

[0038]

[0039] Hereinafter, exemplary implementation examples of the present invention will be described in more detail.

[0040]

[0041] In one aspect, exemplary embodiments of the present invention provide a method for assisting in the diagnosis of osteoporosis, comprising the steps of: measuring plantar pressure of a subject while walking; and predicting the likelihood of being diagnosed with osteoporosis based on the measured plantar pressure.

[0042] The step of measuring plantar pressure during walking can be performed using any method conventional in the art that can measure plantar pressure of an individual during walking, without limitation. For example, it can be performed using a pedobarographic system, but is not limited thereto.

[0043] In one embodiment, the plantar pressure may be the pressure of the 2nd and 3rd metatarsal heads. Here, the 2nd and 3rd metatarsal heads refer to the second and third metatarsal heads.

[0044] In one embodiment, the osteoporosis may be osteoporosis in the lumbar spine or femur region.

[0045] In one embodiment, the likelihood of being diagnosed with osteoporosis can be calculated using the following mathematical formula 1.

[0046] [Mathematical Formula 1]

[0047] = 6.26830-1.86054×(maximum pressure of the 2nd and 3rd metatarsal heads)-0.07031×(body weight)

[0048] (where p is the probability (%) of being diagnosed with osteoporosis).

[0049] In one embodiment, the predicting step is performed such that the maximum pressure of the second and third metatarsal heads of the subject when walking is 2.61 kgf / cm. 2 If the value is less than 0, it may be predicted that the likelihood of being diagnosed with osteoporosis is higher compared to the control group. The control group refers to an individual whose risk of being diagnosed with osteoporosis is compared to the subject. For example, the control group may refer to a normal individual who has never been diagnosed with osteoporosis or has not been diagnosed with osteoporosis, but is not limited thereto.

[0050] In one embodiment, the predicting step is performed such that the maximum pressure of the second and third metatarsal heads of the subject when walking is 2.61 kgf / cm. 2 In this case, it may be predicted that the probability of being diagnosed with osteoporosis is lower compared to the control group. The control group is as described above.

[0051]

[0052] In another aspect, embodiments of the present invention provide a computer program stored in a medium to execute an osteoporosis diagnosis assistance method according to one embodiment of the present invention in combination with hardware.

[0053] That is, the operations or steps by the method described above can be implemented at least partially as a computer program and recorded on a computer-readable recording medium. The computer-readable recording medium on which the program for implementing the operations or steps by the method according to the embodiments is recorded includes all types of recording devices that store data that can be read by a computer. Examples of the computer-readable recording medium include a flash memory type, a hard disk type, a multimedia card micro type, a card-type memory (e.g., SD or XD memory), RAM, ROM, etc. In addition, the computer-readable recording medium can be distributed across network-connected computer systems, so that the computer-readable code can be stored and executed in a distributed manner. In addition, the functional program, code, and code segments for implementing the present embodiment can be easily understood by a person skilled in the art to which the present embodiment belongs. The computer program is composed of a set of commands that can be executed on the above-described computer, and the name thereof may vary depending on the type of the computer. For example, if the form of the computer is a smart phone, the computer program may be referred to as an app.

[0054]

[0055] In another aspect, a device for assisting in the diagnosis of osteoporosis, which executes an assisting method for diagnosing osteoporosis according to embodiments of the present invention, is provided, comprising: a measurement module for measuring plantar pressure when a subject walks; and a prediction module for predicting the possibility of being diagnosed with osteoporosis based on the measured plantar pressure.

[0056] The above measurement module may perform a step of measuring the plantar pressure of the subject while walking. Furthermore, the prediction module may perform a step of predicting the likelihood of being diagnosed with osteoporosis based on the measured plantar pressure. The details of each step have been described above, so a detailed explanation will be omitted.

[0057] In one embodiment, the device may be a smart shoe or a smart insole. The smart shoe or smart insole refers to a wearable device in the form of a shoe or insole. The smart shoe or smart insole may include any device capable of performing the measurement module and the prediction module using methods commonly used in the art, without limitation.

[0058]

[0059] Hereinafter, the present invention will be described in more detail through examples. These examples are intended solely to illustrate the present invention, and it will be apparent to those skilled in the art that the scope of the present invention is not limited by these examples.

[0060]

[0061] Example

[0062]

[0063] [Experimental Method]

[0064] 1. Selection of subjects

[0065]

[0066] From January 2018 to July 2022, patients who visited Seoul National University Bundang Hospital and underwent pedobarography and DEXA (Dual-energy X-ray absorptiometry) at 6-month intervals were selected and enrolled in the study according to the present invention. Pedobarography was performed mainly on patients who visited due to conditions such as hallux valgus (37.5%), flat feet (24.0%), ankle osteoarthritis (14.4%), and ankle instability (4.8%). It is noteworthy that all patients were able to walk without physical limitations or the assistance of assistive devices, regardless of foot or ankle condition. DEXA scanning was mainly performed during routine health assessments as part of the national health screening, and in a small number of patients, it was performed as a clinical evaluation. Specific exclusion criteria were as follows: 1) presence of neuromuscular diseases such as Parkinson's disease, cerebral palsy, or Charcot-Marie-Tooth disease; 2) history of traumatic or insufficient fracture of the lower extremities within the past 6 months; 3) Presence of other conditions that could impair the normal heel-to-toe gait pattern, such as Achilles tendon strain or equinus deformity; 4) Patients with insufficient clinical data, such as body mass index (BMI) or spinal bone density. The age, sex, height, weight, and body mass index (BMI) of the subjects are summarized in Table 1.

[0067]

[0068] 2. Measurement of bone density

[0069] DEXA scans were performed by a qualified technician to reduce errors, and DEXA reports were written by a skilled radiologist. The most significant change in DEXA scan values ​​was 1.2%. All patients underwent scans in the supine position.

[0070] For lumbar spine BMD measurements, patients were instructed to flex their hips and knees to 90 degrees and place them on a cushion. For femoral BMD measurements, patients extended their hips and internally rotated them 15 degrees. As part of the study, BMD was measured at the lumbar spine, left total hip, and left femoral neck (Figure 1). Technicians established each region of interest according to the manufacturer's instructions. Osteoporosis was diagnosed when BMD at the L1-L4 spine, femoral neck, or total femur fell 2.5 standard deviations below the baseline BMD value. In this study, the reference values ​​provided by Hologic (Lee KS et al., 『New Reference Data on Bone Mineral Density 335 and the Prevalence of Osteoporosis in Korean Adults Aged 50 Years or Older: The Korea 336 National Health and Nutrition Examination Survey 2008-2010.』J Korean Med Sci.2014; 29(11):1514-1522. doi: 10.3346 and Je jkms.2014.29.11.1514) were used to diagnose osteoporosis.

[0071]

[0072] 3. Measurement of plantar pressure

[0073] Plantar pressure was measured during comfortable barefoot walking at a self-selected pace using a pedobarogradiographic system (Footwork Pro; AmCube, Berkshire, UK). Following the manufacturer's instructions, participants were instructed to walk naturally along a 4-m-long runway at their usual pace. A pedometer with a seamlessly blended plate and color was positioned at the center of the runway and set at the same height. To acclimate to the setup, participants were encouraged to practice walking the length of the runway several times, completing approximately three to four trials before actual data collection. The mat-shaped pressure plate measured 645 × 520 × 25 mm and contained 4,096 pressure sensors. The active area was 490 × 490 mm, and each sensor measured 7.6 × 7.6 mm. The plate was then coated with polycarbonate. The perceptible pressure range was 10–1200 kPa, and the sampling rate was 100 Hz. Data were transmitted from the sensor array to a desktop computer and analyzed using standard software (Footwork Pro). The measurement sites were divided into eight areas: the hallux, lesser toes, first metatarsal head (M1), second and third metatarsal heads (M2 / 3), fourth and fifth metatarsal heads (M4 / 5), middle foot, medial heel, and lateral heel (Fig. 2a). Dynamic foot pressure measurement of the left foot was used to evaluate the maximum and average pressures for each area of ​​the foot and for the entire foot being measured (Fig. 2b).

[0074]

[0075] 4. Statistics and Analysis

[0076] Descriptive statistics, including means, standard deviations (SD), and proportions, were performed. Data normality was tested using the Kolmogorov-Smirnov test. Means between two groups were compared using the Student's t-test, and proportions between two groups were compared using the chi-square test. Correlations between continuous variables were analyzed using the Pearson correlation coefficient. To assess the influence of body size on the results, correlation tests were repeated using standardized pressure divided by BMI. Univariate regression analysis was performed, and variables with p-values ​​less than 0.1 were selected. Multivariate regression analysis was performed using a parsimonious model. Candidate variables included plantar pressure, which was correlated with BMD, as well as clinical factors such as gender, age, BMI, and comorbidities. For example, regional plantar pressure was selected for the parsimonious model based on its superior explanatory power when it exhibited collinearity with other plantar pressures. Linearity was assessed using scatterplots and trend lines. In addition, clinical factors previously reported to be associated with BMD were incorporated. Furthermore, this regression model achieved a satisfactory R-squared value and demonstrated a low variance inflation factor (VIF). A receiver operating characteristic (ROC) curve was used to identify the optimal plantar pressure criterion for predicting osteoporosis. Furthermore, a multiple logistic regression model was constructed to demonstrate significant discriminatory power.

[0077] The discriminatory ability of the area under the ROC curve (AUC) was assessed as follows: AUC ≥ 0.90 indicates excellent discrimination, 0.80 ≤ AUC < 0.90 indicates good discrimination, 0.70 ≤ AUC < 0.80 indicates fair discrimination, and AUC < 0.70 indicates poor discrimination. A cutoff value for plantar pressure that produced the highest AUC was determined. Plantar pressure values ​​below this threshold indicate an increased risk of osteoporosis. All statistical analyses were performed using R 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria).

[0078]

[0079] [Experimental Results]

[0080] 1. Cohort identification results

[0081] A total of 104 patients were included, and data from the left foot were selected for the final analysis. The mean age of the patients was 62.6 years (SD: 12.4 years), and the mean BMI was 24.7 kg / m 2 (SD: 3.7kg / m 2 ) were. The mean height, weight, and BMI of the 23 men were 166.4 cm (SD: 7.3 cm), 71.0 kg (SD: 11.8 kg), and 25.5 kg / m, respectively. 2 (SD: 2.9 kg / m 2 ) and 81 women had a height of 156.8 cm (SD: 5.5 cm), a weight of 60.1 kg (SD: 9.4 kg), and a body mass of 24.6 kg / m, respectively. 2 (SD: 3.9kg / m 2 ) was 0.98 g / cm2. The average bone density of the lumbar spine, femoral neck, and total femur was 0.98 g / cm2, respectively. 2 (SD: 0.18 g / cm 2 ), 0.72 g / cm 2 (SD: 0.13 g / cm 2 ) and 0.85 g / cm 2(SD: 0.14 g / cm 2 ) was shown. The maximum pressure of the entire foot during walking was 1.41 kgf / cm. 2 (SD: 0.33 kgf / cm 2 ) (Table 1).

[0082]

[0083]

[0084]

[0085] 2. Plantar pressure in eight areas of the sole stratified by the presence or absence of osteoporosis

[0086] The plantar pressures at each site are shown in Table 2. When stratified by the presence or absence of osteoporosis, the maximum and mean pressures of the second and third metatarsal heads showed statistically significant differences between the groups.

[0087]

[0088]

[0089]

[0090] 3. Correlation between bone density and plantar pressure and application of standardization by BMI

[0091] Lumbar spine BMD showed a significant correlation with the maximum pressure of the 2nd and 3rd, 4th and 5th metatarsal heads, and the average pressure of the 4th and 5th metatarsal heads. Femoral neck BMD showed a significant correlation with the maximum pressure of the 2nd and 3rd metatarsal heads, 4th and 5th metatarsal heads, and the medial heel. In addition, a correlation was observed with the average pressure of the great toe, 2nd and 3rd metatarsal heads, 4th and 5th metatarsal heads, and the medial heel. Total femoral BMD showed a significant correlation with the maximum pressure of the 2nd and 3rd, 4th and 5th metatarsal heads, and the average pressure of the great toe, 2nd and 3rd metatarsal heads, 4th and 5th metatarsal heads, and the medial heel (Tables 3 and 4).

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] After normalization by BMI, lumbar spine BMD was significantly correlated with the standardized peak pressures of the 2nd and 3rd, 4th and 5th metatarsal heads, and the standardized mean pressures of the 4th and 5th metatarsal heads. Femoral neck BMD was significantly correlated with the standardized peak pressures of the 2nd and 3rd metatarsal heads, 4th and 5th metatarsal heads, and medial heel. In addition, correlations were observed with the normalized mean pressures of the great toe, 2nd and 3rd metatarsal heads, and 4th and 5th metatarsal heads. Total femoral BMD was significantly correlated with the standardized peak pressures of the 2nd and 3rd, 4th and 5th metatarsal heads. It was correlated with the normalized mean pressures of the 2nd and 3rd metatarsal heads and 4th and 5th metatarsal heads (Tables 5 and 6).

[0098]

[0099]

[0100]

[0101]

[0102]

[0103] 4. Multivariate linear regression results

[0104] Following the univariate regression analysis (Tables 7 to 9), multivariate regression analysis using a parsimonious model yielded the following results.

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112] For lumbar bone density, the average pressure of the 4th and 5th metatarsal heads showed a significant correlation (p=0.041, adjusted R 2 =0.081, Table 10). Femoral neck BMD was significantly associated with age (p=0.049), body weight (p=0.044), and peak pressure of the second and third metatarsal heads (p=0.002, adjusted R 2 =0.213, Table 11) had a significant correlation.

[0113]

[0114]

[0115]

[0116]

[0117]

[0118] Total femoral bone mineral density was significantly correlated with sex (p=0.004), age (p=0.022), weight (p=0.013), rheumatoid arthritis (p=0.004), and peak pressure of the second and third metatarsal heads (p=0.003, adjusted R 2 =0.360, Table 12).

[0119]

[0120]

[0121]

[0122] 5. Osteoporosis and ROC curve analysis

[0123] Of the nine patients with osteoporosis, three were diagnosed only in the lumbar spine and three in the femur. In the logistic regression model for the diagnosis of osteoporosis, the maximum pressure of the second and third metatarsal heads showed the highest AUC of 0.819. Furthermore, the pressure there was 2.61 kgf / cm 2 When the value was less than 0.001, the specificity for osteoporosis diagnosis was 0.684, and the sensitivity was 0.889. Furthermore, plantar pressure exhibited a high negative predictive value of 98.5% (Fig. 3a). Therefore, the model combining the weights of the ROC analysis showed an improved AUC of 0.833 (Fig. 4).

[0124] Finally, using the maximum pressure of the second and third metatarsal heads and body weight, which are variables with high utility as negative predictors of osteoporosis diagnosis, the regression analysis equation (Mathematical Equation 1 below) of multivariate logistic regression analysis using the R function was obtained.

[0125] [Mathematical Formula 1]

[0126] = 6.26830-1.86054×(maximum pressure of the 2nd and 3rd metatarsal heads)-0.07031×(body weight)

[0127] (where p is the probability (%) of being diagnosed with osteoporosis).

[0128] Therefore, it can be confirmed that the possibility of effectively diagnosing osteoporosis can be predicted using the maximum pressure of the second and third metatarsal heads using a method according to one embodiment of the present invention.

Claims

1. A step of measuring the plantar pressure of the subject while walking; and An osteoporosis diagnosis assistance method, comprising: a step of predicting the possibility of being diagnosed with osteoporosis based on the measured plantar pressure.

2. In paragraph 1, The above plantar pressure is an auxiliary method for diagnosing osteoporosis, which is the pressure of the 2nd and 3rd metatarsal heads.

3. In paragraph 1, The above osteoporosis is osteoporosis in the lumbar spine or femur region, and is an auxiliary method for diagnosing osteoporosis.

4. In paragraph 1, A method characterized in that the likelihood of being diagnosed with the above osteoporosis is calculated using the following mathematical formula 1: [Mathematical Formula 1] = 6.26830-1.86054×(maximum pressure of the 2nd and 3rd metatarsal heads)-0.07031×(body weight) (where p is the probability (%) of being diagnosed with osteoporosis).

5. In paragraph 1, The above-mentioned predicting step is that the maximum pressure of the 2nd and 3rd metatarsal heads when the subject walks is 2.61 kgf / cm 2 An osteoporosis diagnosis assistance method that predicts that, if less than, the probability of being diagnosed with osteoporosis is higher than that of the control group.

6. In paragraph 1, The above-mentioned predicting step is that the maximum pressure of the 2nd and 3rd metatarsal heads when the subject walks is 2.61 kgf / cm 2 An osteoporosis diagnosis assistance method that predicts that, in the above cases, the probability of being diagnosed with osteoporosis is lower compared to the control group.

7. A computer program stored on a medium that is combined with hardware and executes any one of the methods of claims 1 to 6.

8. An osteoporosis diagnosis auxiliary device that performs the method of any one of clauses 1 to 6, A measurement module for measuring the plantar pressure of a subject when walking; and A device comprising a prediction module for predicting the possibility of being diagnosed with osteoporosis based on the measured plantar pressure.

9. In paragraph 8, The above device is a smart shoe or a smart insole.

Images