Continuous Gait Phase Estimation for Multi-Locomotion Tasks Using Ground Reaction Force Data.

Existing studies on gait phase estimation generally involve walking experiments using inertial measurement units under limited walking conditions (WCs). In this study, a gait phase estimation algorithm is proposed that uses data from force sensing resistors (FSRs) and a Bi-LSTM model. The proposed algorithm estimates gait phases in real time under various WCs, e.g., walking on paved/unpaved roads, ascending and descending stairs, and ascending or descending on ramps. The performance of the proposed algorithm is evaluated by performing walking experiments on ten healthy adult participants. An average gait estimation accuracy exceeding 90% is observed with a small error (root mean square error = 0.794, R2 score = 0.906) across various WCs. These results demonstrate the wide applicability of the proposed gait phase estimation algorithm using various insole devices, e.g., in walking aid control, gait disturbance diagnosis in daily life, and motor ability analysis.

A framework of insole blanking robot based on adaptive edge detection and FSPS-BIT* path planning.

Insole blanking production technology plays a vital role in contemporary machining and manufacturing industries. Existing insole blanking production models have limitations because most robots are required to accurately position the workpiece to a predetermined location, and special auxiliary equipment is usually required to ensure the precise positioning of the robot. In this paper, we present an adaptive blanking robotic system for different lighting environments, which consists of an industrial robot arm, an RGB-D camera configuration, and a customized insole blanking table and mold. We introduce an innovative edge detection framework that utilizes color features and morphological parameters optimized through particle swarm optimization (PSO) techniques to Adaptive recognition of insole edge contours. A path planning framework based on FSPS-BIT* is also introduced, which integrates the BIT* algorithm with the FSPS algorithm for efficient path planning of the robotic arm.

Perception of Running Shoe Cushioning: Determining the Correspondence between Subjective Assessment and In-vitro Measurement.

Running shoes, and in particular insoles, are the first interface between runners and running surface. Different insole attenuation properties may vary perception of cushioning and, accordingly, the effect on muscle adaptation. The aim of this study is to find the just noticeable difference between four insole materials, and investigate the parameters of in-vitro measurement of impact testing to predict cushioning comfort. Nineteen (n = 19) male participants were recruited from the sports center at the Technical University of Munich with a mean age of 23.89 (SD = 2.31), weight of 73.52 kg (SD =3.08), and height 178.84 cm (SD =2.81). Four insole samples, one with the highest peak acceleration (EPDM =17.9g), one with the lowest (S.Tk = 8.3g) and the two materials with middle range magnitudes (IP.GL= 11.5g and S.Tn = 12.2g), were selected to use in the subjective measurement. We used the impact testing method to evaluate the in-vitro physical properties of insoles in running shoes. In addition, two parameters of peak acceleration were measured as follows: Jolt α was calculated at a slope of between 5-20 % of inertial impact force and Jolt ß was calculated at a slope of between 0-88 Newtons of inertial impact force. Participants performed six pairwise comparison tests with shoes which were equipped with one of the four insoles in a random order. A minimum 6% increase in cushioning properties, notably between 11.5g (S.Tn) and 12.2g (IP.Gl), was discerned through the paired tests. In simpler terms, participants were able to detect a mere 0.7g as the just noticeable difference. In addition, our findings revealed that an increase of Jolt α and Jolt ß resulted in a reduction in perception of comfort. There was a negative and significant correlation between Jolt α and perceived cushioning and, similarly, between Jolt ß and perceived cushioning r (10) = -0.93, p = 0.00001. No correlation was found between peak acceleration and cushioning comfort (p = 0.1). These discoveries may facilitate a better understanding of how human adaptation can occur with different cushioning.

The Effect of Dimple Insole Design on the Plantar Temperature and Pressure in People with Diabetes and in Healthy Individuals.

An increase in plantar pressure and skin temperature is commonly associated with an increased risk of diabetic foot ulcers. However, the effect of insoles in reducing plantar temperature has not been commonly studied. The aim was to assess the effect of walking in insoles with different features on plantar temperature. Twenty-six (F/M:18/8) participants-13 with diabetes and 13 healthy, aged 55.67 ± 9.58 years-participated in this study. Skin temperature at seven plantar regions was measured using a thermal camera and reported as the difference between the temperature after walking with an insole for 20 m versus the baseline temperature. The mixed analyses of variance indicated substantial main effects for the Insole Condition, for both the right [Wilks' Lambda = 0.790, F(14, 492) = 4.393, p < 0.01, partial eta squared = 0.111] and left feet [Wilks' Lambda = 0.890, F(14, 492) = 2.103, p < 0.011, partial eta squared = 0.056]. The 2.5 mm-tall dimple insole was shown to be significantly more effective at reducing the temperature in the hallux and third met head regions compared to the 4 mm-tall dimple insole. The insoles showed to be significantly more effective in the diabetes group versus the healthy group, with large effect size for the right [Wilks' Lambda = 0.662, F(14, 492) = 8.037, p < 0.000, Partial eta-squared = 0.186] and left feet [Wilks' Lambda = 0.739, F(14, 492) = 5.727, p < 0.000, Partial eta-squared = 0.140]. This can have important practical implications for designing insoles with a view to decrease foot complications in people with diabetes.

High-Output Triboelectric Nanogenerator Achieved through Conductive Layer Strategy for Motion Step Sensing.

As an emerging high-efficiency energy conversion device, improving the output of triboelectric nanogenerators (TENGs) is still a key method to promote practical application of TENGs. This paper systematically investigated the influence of component composition, thickness, and surface morphology of the metal conducting layer on the performance of triboelectric nanogenerators. It has been established that these three factors have a significant influence on the output performance of TENGs. Among the four common metals Au, Pt, Ag, and Cu, the triboelectric nanogenerator achieves its maximum output when utilizing Ag as the conducting layer, with optimal performance observed at a thickness of 278 nm. TENGs with nanostructured conducting layers have better output as the nanostructure amplifies the induction charging area, thereby effectively augmenting the performance of TENGs. In particular, when contrasted with a triboelectric nanogenerator utilizing copper foil as the conducting layer alongside poly(vinylidene difluoride) and Nylon-11 as friction layers in the common work, the short-circuit current of the triboelectric nanogenerator increased by 2.3 times, and the maximum short-circuit current reached 149 µA when the conducting layer was replaced with Ag, and the enhanced triboelectric nanogenerator successfully illuminated 1536 commercial LEDs. In addition, the TENG-based smart insoles combined with pedometers can realize signal sensing and the real-time recording of steps during exercise. This research provides a new simple and reliable method to further improve the output of the TENG.

Smart insole-based analysis of gait biomechanics for insoles in patients with flatfoot.

BACKGROUND: Assessing the effect of insoles on gait biomechanics and foot comfort remains challenging. Our novel in-insole-type wearable sensor device (smart insole) enables accurate quantitative evaluation of gait parameters without affecting the subject's foot comfort. RESEARCH QUESTION: What are the effects of insoles on gait biomechanics and foot comfort in patients with flatfoot, as evaluated using a novel smart insole? METHODS: Thirty-three subjects with 61 flatfeet were recruited. Three different types of prefabricated insoles were tested: a control insole as an experimental control, a flat insole with only cushion pads for the shock absorbing function, and an arch support insole with both cushioning pads and arch support functions. Gait parameters and visual analog scale (VAS) scores for foot comfort were measured during 30 m of straight walking with each insole incorporating the wearable sensor device. The differences in gait parameters and foot comfort between the flat and arch support insoles relative to the control insole were analyzed. Additionally, the correlations between gait parameters and foot comfort were evaluated. RESULTS: Maximum plantarflexion angle significantly decreased (p = 0.03) and the toe-out angle significantly increased (p < 0.01) with arch support insoles compared to flat insoles. Significantly better foot comfort was demonstrated when walking with arch support insoles than with flat insoles (p < 0.01). The only gait parameter correlated with foot comfort was foot lift height while walking with an arch support insole (r = -0.45, p < 0.01). SIGNIFICANCE: A novel smart insole revealed that foot lift height was a key gait parameter for determining foot comfort while walking with an arch support insole. Our findings provide important evidence for selecting a comfortable flatfoot insole based on gait data measured using a smart insole.

Consumer-priced wearable sensors combined with deep learning can be used to accurately predict ground reaction forces during various treadmill running conditions.

Ground reaction force (GRF) data is often collected for the biomechanical analysis of running, due to the performance and injury risk insights that GRF analysis can provide. Traditional methods typically limit GRF collection to controlled lab environments, recent studies have looked to combine the ease of use of wearable sensors with the statistical power of machine learning to estimate continuous GRF data outside of these restrictions. Before such systems can be deployed with confidence outside of the lab they must be shown to be a valid and accurate tool for a wide range of users. The aim of this study was to evaluate how accurately a consumer-priced sensor system could estimate GRFs whilst a heterogeneous group of runners completed a treadmill protocol with three different personalised running speeds and three gradients. Fifty runners (25 female, 25 male) wearing pressure insoles made up of 16 resistive sensors and an inertial measurement unit ran at various speeds and gradients on an instrumented treadmill. A long short term memory (LSTM) neural network was trained to estimate both vertical ( G R F v ) and anteroposterior ( G R F a p ) force traces using leave one subject out validation. The average relative root mean squared error (rRMSE) was 3.2% and 3.1%, respectively. The mean ( G R F v ) rRMSE across the evaluated participants ranged from 0.8% to 8.8% and from 1.3% to 17.3% in the ( G R F a p ) estimation. The findings from this study suggest that current consumer-priced sensors could be used to accurately estimate two-dimensional GRFs for a wide range of runners at a variety of running intensities. The estimated kinetics could be used to provide runners with individualised feedback as well as form the basis of data collection for running injury risk factor studies on a much larger scale than is currently possible with lab based methods.

Detection of freezing of gait in Parkinson's disease from foot-pressure sensing insoles using a temporal convolutional neural network.

Backgrounds: Freezing of gait (FoG) is a common and debilitating symptom of Parkinson's disease (PD) that can lead to falls and reduced quality of life. Wearable sensors have been used to detect FoG, but current methods have limitations in accuracy and practicality. In this paper, we aimed to develop a deep learning model using pressure sensor data from wearable insoles to accurately detect FoG in PD patients. Methods: We recruited 14 PD patients and collected data from multiple trials of a standardized walking test using the Pedar insole system. We proposed temporal convolutional neural network (TCNN) and applied rigorous data filtering and selective participant inclusion criteria to ensure the integrity of the dataset. We mapped the sensor data to a structured matrix and normalized it for input into our TCNN. We used a train-test split to evaluate the performance of the model. Results: We found that TCNN model achieved the highest accuracy, precision, sensitivity, specificity, and F1 score for FoG detection compared to other models. The TCNN model also showed good performance in detecting FoG episodes, even in various types of sensor noise situations. Conclusions: We demonstrated the potential of using wearable pressure sensors and machine learning models for FoG detection in PD patients. The TCNN model showed promising results and could be used in future studies to develop a real-time FoG detection system to improve PD patients' safety and quality of life. Additionally, our noise impact analysis identifies critical sensor locations, suggesting potential for reducing sensor numbers.

Effect of different inner pressures of air insoles and walking durations on plantar pressure time integral.

Air insoles have provided insights for reducing the risk of diabetic foot ulcers (DFU). The pressure time integral (PTI) is an effective assessment that considers the time effect in various physical activities. We investigated the interactions between three different insole inner pressures (80, 160, and 240 mmHg) and two walking durations (10 and 20 min). The big toe (T1), first metatarsal head (M1), and second metatarsal head (M2) were investigated in 13 healthy participants. One-way analysis of variance (ANOVA) showed that the effects of each insole inner pressure significantly differed (P < 0.05) with a 10 min walking duration. The PTI values resulting from 80 mmHg in M2 (38.4 ± 3.8, P = 0.002) and 160 mmHg in M1 (44.3 ± 4.3, P = 0.027) were lower than those from 240 mmHg. Additionally, the paired t test showed that the effects of each walking duration were also considerably different at 160 mmHg. The PTI at 10 min was lower than that at 20 min in M1 (44.31 ± 4.31, P = 0.015) and M2 (47.14 ± 5.27, P = 0.047). Thus, we suggest that walking with a pressure of 160 mmHg for 10 min has a lower risk of DFU.

Comparison of the Accuracy of Ground Reaction Force Component Estimation between Supervised Machine Learning and Deep Learning Methods Using Pressure Insoles.

The three Ground Reaction Force (GRF) components can be estimated using pressure insole sensors. In this paper, we compare the accuracy of estimating GRF components for both feet using six methods: three Deep Learning (DL) methods (Artificial Neural Network, Long Short-Term Memory, and Convolutional Neural Network) and three Supervised Machine Learning (SML) methods (Least Squares, Support Vector Regression, and Random Forest (RF)). Data were collected from nine subjects across six activities: normal and slow walking, static with and without carrying a load, and two Manual Material Handling activities. This study has two main contributions: first, the estimation of GRF components (Fx, Fy, and Fz) during the six activities, two of which have never been studied; second, the comparison of the accuracy of GRF component estimation between the six methods for each activity. RF provided the most accurate estimation for static situations, with mean RMSE values of RMSE_Fx = 1.65 N, RMSE_Fy = 1.35 N, and RMSE_Fz = 7.97 N for the mean absolute values measured by the force plate (reference) RMSE_Fx = 14.10 N, RMSE_Fy = 3.83 N, and RMSE_Fz = 397.45 N. In our study, we found that RF, an SML method, surpassed the experimented DL methods.

Validity and Test-Retest Reliability of Spatiotemporal Running Parameter Measurement Using Embedded Inertial Measurement Unit Insoles.

Running is the basis of many sports and has highly beneficial effects on health. To increase the understanding of running, DSPro® insoles were developed to collect running parameters during tasks. However, no validation has been carried out for running gait analysis. The aims of this study were to assess the test-retest reliability and criterion validity of running gait parameters from DSPro® insoles compared to a motion-capture system. Equipped with DSPro® insoles, a running gait analysis was performed on 30 healthy participants during overground and treadmill running using a motion-capture system. Using an intraclass correlation coefficient (ICC), the criterion validity and test-retest reliability of spatiotemporal parameters were calculated. The test-retest reliability shows moderate to excellent ICC values (ICC > 0.50) except for propulsion time during overground running at a fast speed with the motion-capture system. The criterion validity highlights a validation of running parameters regardless of speeds (ICC > 0.70). This present study validates the good criterion validity and test-retest reliability of DSPro® insoles for measuring spatiotemporal running gait parameters. Without the constraints of a 3D motion-capture system, such insoles seem to be helpful and relevant for improving the care management of active patients or following running performance in sports contexts.

Immediate effects of insoles applied to the sound side lower extremity of patients with chronic hemiplegia during walking.

Background: Asymmetric gait patterns are mostly observed in hemiplegic stroke patients. These abnormal gait patterns resulting in abnormal speed, and decreased ability in daily of activity living. Objective: This study aimed to determine the immediate changes in gait parameters and plantar pressure during elevation by wearing an insole on the sound side lower extremity of patients with hemiplegia. Methods: Thirty-six participants were recruited, comprising those with a post-stroke follow-up of ≥3 months and a functional ambulation category score of ≥2. The participants were asked to walk with and without a 1 cm insole in the shoe of their sound side, and the order of wearing or not wearing the insole was randomized. Gait parameters, bilateral gait parameters, and dynamic plantar pressure were measured using the GAITRite Walkway System. Results: Paired t-test was used to examine immediate changes in gait parameters and plantar pressure with and without insoles during walking in the same group. Overall, gait velocity and step length significantly decreased (p < 0.05), whereas step time significantly increased (p < 0.05). The swing phase of the affected sidelower extremities significantly increased (p < 0.05), and the stance phase significantly decreased (p < 0.05). Double-support unloading phase (pre-swing phase) significantly increased (p < 0.05). The changes in plantar pressure were significantly increased in some lateral zones and significantly decreased in the medial zone of the mid-hindfoot, both in terms of pressure per time and peak pressure (p < 0.05). Conclusion: Although this study did not show immediate positive effects on gait parameters and gait cycle, it is expected that sensory input from the sole of the foot through changes in plantar pressure may help improve gait asymmetry and regulate postural symmetry.

Comparative study of the effects of custom-made insole and ordinary insole in adults with flexible flatfoot on different slopes.

BACKGROUND: Flatfoot (pes planus) is a common foot deformity, and its causes are mainly related to age, gender, weight, and genetics. Previous studies have shown that custom-made insoles could have a positive effect in improving plantar pressure and symptoms in individuals with flexible flatfeet, but it remains to be explored whether they can still show benefits in daily walking on different slopes. OBJECTIVE: This study aims to investigate a custom-made insole based on plantar pressure redistribution and to verify its effectiveness by gait analysis on different slopes. METHODS: We recruited 10 subjects and compared the peak pressure and impulse in each area between custom-made insole (CI) and ordinary insole (OI) groups. RESULTS: The results illustrate that CI raises the pressure in T area, improves the ability of the subjects to move forward in the slope walking, which was beneficial to gait stability. CONCLUSION: The redistribution of pressure in MF and MH area is promoted to provide active protection for subjects. Meanwhile, CI could decrease the impulse in MF area during uphill and level walking, which effectively reduces the accumulation of fatigue during gait. Moreover, avoiding downhill walking could be able to protect foot from injury in daily life.

Deep-learning enabled smart insole system aiming for multifunctional foot-healthcare applications.

Real-time foot pressure monitoring using wearable smart systems, with comprehensive foot health monitoring and analysis, can enhance quality of life and prevent foot-related diseases. However, traditional smart insole solutions that rely on basic data analysis methods of manual feature extraction are limited to real-time plantar pressure mapping and gait analysis, failing to meet the diverse needs of users for comprehensive foot healthcare. To address this, we propose a deep learning-enabled smart insole system comprising a plantar pressure sensing insole, portable circuit board, deep learning and data analysis blocks, and software interface. The capacitive sensing insole can map both static and dynamic plantar pressure with a wide range over 500 kPa and excellent sensitivity. Statistical tools are used to analyze long-term foot pressure usage data, providing indicators for early prevention of foot diseases and key data labels for deep learning algorithms to uncover insights into the relationship between plantar pressure patterns and foot issues. Additionally, a segmentation method assisted deep learning model is implemented for exercise-fatigue recognition as a proof of concept, achieving a high classification accuracy of 95%. The system also demonstrates various foot healthcare applications, including daily activity statistics, exercise injury avoidance, and diabetic foot ulcer prevention.

Effects of individually optimized rocker midsoles and self-adjusting insoles on dynamic stability in persons with diabetes mellitus and neuropathy.

BACKGROUND: Persons with diabetic peripheral neuropathy (DPN) may face challenges such as balance issues due to reduced somatosensory feedback and an increased risk of developing diabetic foot ulcers (DFUs) due to increased plantar pressure. Pressure reducing footwear is thought to further impair balance. We introduced 3D-printed rocker midsoles and self-adjusting insoles that are able to reduce elevated plantar pressure values and aimed to prevent balance deterioration. However, their effect on the balance during walking (dynamic stability) is not analyzed yet. RESEARCH QUESTION: Is dynamic stability of persons with DPN impaired compared to healthy individuals and what is the effect of the 3D-printed rocker midsoles and self-adjusting insoles on the dynamic stability in this population? METHODS: Dynamic stability, specifically the margins of stability (MOS) in the anterior-posterior (AP) and medio-lateral (ML) direction, was measured in ten healthy and nineteen persons with DPN. Independent-samples t-test was applied to analyze the difference in the MOS between groups. One-way repeated measures analyses of variance (ANOVA) was conducted to test the difference between the therapeutic footwear combinations within the DPN group. RESULTS: There is no significant difference between the healthy and DPN group in MOS-AP. MOS-ML is significantly larger in DPN compared to the healthy participants. Using the self-adjusting insole shows a significantly lower (negative) MOS-AP compared to when using a rocker shoe within the DPN group. SIGNIFICANCE: This study provides valuable information on whether DPN and our therapeutic footwear have a negative effect on the dynamic stability. DPN does not have a negative effect on dynamic stability in the AP direction. For the ML direction, DPN seems to cause larger MOS-ML by likely using a compensation strategy (e.g., wider steps) while our experimental footwear does not further impair the MOS-ML.

Gait Pattern Analysis: Integration of a Highly Sensitive Flexible Pressure Sensor on a Wireless Instrumented Insole.

Gait phase monitoring wearable sensors play a crucial role in assessing both health and athletic performance, offering valuable insights into an individual's gait pattern. In this study, we introduced a simple and cost-effective capacitive gait sensor manufacturing approach, utilizing a micropatterned polydimethylsiloxane dielectric layer placed between screen-printed silver electrodes. The sensor demonstrated inherent stretchability and durability, even when the electrode was bent at a 45-degree angle, it maintained an electrode resistance of approximately 3 Ω. This feature is particularly advantageous for gait monitoring applications. Furthermore, the fabricated flexible capacitive pressure sensor exhibited higher sensitivity and linearity at both low and high pressure and displayed very good stability. Notably, the sensors demonstrated rapid response and recovery times for both under low and high pressure. To further explore the capabilities of these new sensors, they were successfully tested as insole-type pressure sensors for real-time gait signal monitoring. The sensors displayed a well-balanced combination of sensitivity and response time, making them well-suited for gait analysis. Beyond gait analysis, the proposed sensor holds the potential for a wide range of applications within biomedical, sports, and commercial systems where soft and conformable sensors are preferred.

Forensic measurement comparison of foot insole impressions.

In forensic podiatry, footprints have been shown to provide a valuable source of discriminatory information. Footprints may be found in various forms, such as bare footprints, sock-clad footprints, or as impressions on insoles within footwear. This study utilized quantitative measures of foot impressions on pairs of insoles from shoes worn by the same person from a population of 31 adults. The measurements were determined by using the Reel method and comprised measurements from the heel to the tips of the toes and width of the ball. The purpose of the study was to assess the margin of error for these measurements to determine whether they were sufficiently accurate for forensic use. A secondary purpose of this study was to determine whether the analyst's experience or lack thereof in forensic podiatry had an impact on the precision of measurement data. The insole foot impressions were assessed by two podiatrists with forensic podiatry experience in footprint analysis, footprint research, and in using the Reel method of footprint measurement, as well as by three students of podiatric medicine without any such experience. A statistical analysis of the data from the study was performed using SPSS v28 (IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp). The most reliable measurements were of forefoot width, heel to first toe, heel to second toe, and heel to fourth toe. The greatest variation occurred in the measurements of the heel to the third and fifth toes. The measurements of the forensic podiatrist analysts showed less variability than those of the podiatry students, suggesting that measurement precision is related to the experience of the analyst.

Role of insole material in treatment of plantar fasciitis: A randomized clinical trial.

BACKGROUND: In this randomized clinical trial, we compared the early effects of polyethylene (PE), polyurethane (PU), and Carbon Fiber insoles in the treatment of PF using a set of patient-reported outcomes. METHODS: Patients were randomly allocated one of the three prefabricated insoles - Carbon Fiber (n = 14), PU (n = 14), or PE (n = 17) for regular use. Their response was recorded using PROMIS 3a (for pain intensity), PROMIS 4a (for pain interference), FAOS (Foot and Ankle Outcome Score), and VAS for pain at baseline, two, six, and twelve weeks. RESULTS: The PROMIS pain intensity scores improved in both the Carbon Fiber and the PE groups starting at the 6th week (p = 0.04) and 2nd week (p = 0.002), respectively. PROMIS pain interference scores also showed positive trends in these two groups (p = 0.02, p = 0.004, respectively). CONCLUSION: Prefabricated Carbon Fiber and PE insoles showed significant pain-reducing effects in patients with PF. LEVELS OF EVIDENCE: Level I, Randomized controlled trial.

Smart wearable insoles in industrial environments: A systematic review.

BACKGROUND AND PURPOSE: Industrial environments present unique challenges in ensuring worker safety and optimizing productivity. The emergence of smart wearable technologies such as smart insoles has provided new opportunities to address these challenges through accurate unobtrusive monitoring and analysis of workers' activities and physical parameters. This systematic review aims to analyze the utilization of smart wearable insoles in industrial environments, focusing on their applications, employed analysis methods, and potential future directions. METHODS: A comprehensive review was conducted, involving the analysis of 27 papers that utilized smart wearable insoles in industrial settings. The reviewed articles were evaluated to determine the trends in application and methodology, explore the implementation of smart insoles across different industries, and identify the prevalent machine learning models and analyzed activities in the relevant literature. RESULTS: The majority of the reviewed articles (67%) primarily focused on human activity recognition and gesture estimation using smart wearable insoles, aiming to enhance safety and productivity in industrial settings. Furthermore, 10% of the studies focused on fatigue identification, 10% on slip, trip, and fall hazard detection, and 13% on biomechanical analyses of workers' body joint loads. The construction industry accounted for approximately 60% of the studies conducted in industrial settings using smart insoles. The most prevalent machine learning models utilized in these studies were neural networks (48%), support vector machines (33%), k-nearest neighbors (30%), decision trees (26%), and random forests (15%). These models achieved median accuracies of 95%, 96%, 91%, 92%, and 95%, respectively. Among the analyzed activities, walking, bending with/without lifting/lowering a load, and carrying a load were the most frequently considered, with frequencies of 10, 10, and 7 out of the 27 studies, respectively. CONCLUSION: The findings of this systematic review demonstrate the growing interest in implementing smart wearable insoles in industrial environments to enhance safety and productivity. However, the effectiveness of these systems is dependent on factors such as accuracy, reliability, and generalizability of the models. The review highlights the need for further research to address these challenges and to explore the potential of these systems for use in other industrial applications such as manufacturing. Overall, this systematic review provides valuable insights for researchers, practitioners, and policymakers in the field of occupational health and safety.