Random forest algorithms to classify frailty and falling history in seniors using plantar pressure measurement insoles: a large-scale feasibility study.

BACKGROUND: Frailty and falls are two adverse characteristics of aging that impair the quality of life of senior people and increase the burden on the healthcare system. Various methods exist to evaluate frailty, but none of them are considered the gold standard. Technological methods have also been proposed to assess the risk of falling in seniors. This study aims to propose an objective method for complementing existing methods used to identify the frail state and risk of falling in older adults. METHOD: A total of 712 subjects (age: 71.3 ± 8.2 years, including 505 women and 207 men) were recruited from two Japanese cities. Two hundred and three people were classified as frail according to the Kihon Checklist. One hundred and forty-two people presented with a history of falling during the previous 12 months. The subjects performed a 45 s standing balance test and a 20 m round walking trial. The plantar pressure data were collected using a 7-sensor insole. One hundred and eighty-four data features were extracted. Automatic learning random forest algorithms were used to build the frailty and faller classifiers. The discrimination capabilities of the features in the classification models were explored. RESULTS: The overall balanced accuracy for the recognition of frail subjects was 0.75 ± 0.04 (F1-score: 0.77 ± 0.03). One sub-analysis using data collected for men aged > 65 years only revealed accuracies as high as 0.78 ± 0.07 (F1-score: 0.79 ± 0.05). The overall balanced accuracy for classifying subjects with a recent history of falling was 0.57 ± 0.05 (F1-score: 0.62 ± 0.04). The classification of subjects relative to their frailty state primarily relied on features extracted from the plantar pressure series collected during the walking test. CONCLUSION: In the future, plantar pressures measured with smart insoles inserted in the shoes of senior people may be used to evaluate aspects of frailty related to the physical dimension (e.g., gait and balance alterations), thus allowing assisting clinicians in the early identification of frail individuals.

Effect of gait protocols and postoperative shoes on off-loading of forefoot in preoperative patients for forefoot disorders.

OBJECTIVES: The purpose of this study was to clarify the effect of gait protocols and postoperative shoes on forefoot load in preoperative patients for forefoot disorders and compare footwear comfort between different types of postoperative shoes. METHODS: Fourteen subjects scheduled to undergo forefoot surgeries were recruited. The maximum force under the forefoot region was measured during 10 m straight walking in two gait patterns with six different shoe types. Visual analogue scale (VAS) scores for footwear comfort, subjective lower thigh pain, and electrical activities of lower thigh muscles were also evaluated. RESULTS: The body weight-normalized maximum force under the forefoot region significantly decreased in step-to gait compared to normal gait regardless of the shoe types used. Under the same gait condition, no significant difference was observed in the forefoot off-loading effect between the different shoe types used. Significantly worse VAS scores, significantly higher tibialis anterior muscle activities, and complaints of lower thigh pain were demonstrated in the gait with the reverse camber shoe. CONCLUSIONS: Gait protocol of step-to gait had more forefoot off-loading effect than postoperative shoes. The forefoot off-loading effect did not differ among the postoperative shoes, suggesting that postoperative shoes can be selected with an emphasis on footwear comfort.

A 'Giving way' captured during walking by gyroscopes and plantar force sensors.

'Giving way' is a rapid inversion of the rear part of the foot, which does not result in an acute lateral ankle sprain. It is a specific event for patients with chronic ankle instability (CAI). This report described a 'giving way' in an 18-year-old female with CAI, which was biomechanically captured during walking on the flat surface. Shoes with unstable heel having a hemisphere rubber on the sole of the heel were used to provoke 'giving way'. To the authors' knowledge, this is the first report describing the kinetics of 'giving way' during walking. The event of 'giving way' was captured by an Inertial Motor Unit located on the dorsum of the foot and an insole-shaped plantar force measurement device. 'Giving way' provided distinctive data on both kinds of devices. Gyroscope data showed a rapid increase of inversion/plantarflexion/internal rotation (maximum levels: 204/280/346 degree/s) and following eversion/dorsiflexion/external rotation (maximum levels: 509/798/797 degree/s) of the foot segment at 350 ms - 492 ms after the heel strike. Plantar force data demonstrated the rapid decrease and subsequent recovery of the regional force on the head of 1st metatarsal head, suggesting a rapid inversion followed by the foot's defensive eversion. Although the maximum angular velocity of the inversion was smaller and the duration of inversion phase of 'giving way' was shorter than in previous reports, the characteristics of the following eversion phase of 'giving way' were consistent with earlier reports. The eversion must be a more specific phase than the inversion in the kinematics of 'giving way'.

Random forest algorithms for recognizing daily life activities using plantar pressure information: a smart-shoe study.

BACKGROUND: Wearable activity trackers are regarded as a new opportunity to deliver health promotion interventions. Indeed, while the prediction of active behaviors is currently primarily relying on the processing of accelerometer sensor data, the emergence of smart clothes with multi-sensing capacities is offering new possibilities. Algorithms able to process data from a variety of smart devices and classify daily life activities could therefore be of particular importance to achieve a more accurate evaluation of physical behaviors. This study aims to (1) develop an activity recognition algorithm based on the processing of plantar pressure information provided by a smart-shoe prototype and (2) to determine the optimal hardware and software configurations. METHOD: Seventeen subjects wore a pair of smart-shoe prototypes composed of plantar pressure measurement insoles, and they performed the following nine activities: sitting, standing, walking on a flat surface, walking upstairs, walking downstairs, walking up a slope, running, cycling, and completing office work. The insole featured seven pressure sensors. For each activity, at least four minutes of plantar pressure data were collected. The plantar pressure data were cut in overlapping windows of different lengths and 167 features were extracted for each window. Data were split into training and test samples using a subject-wise assignment method. A random forest model was trained to recognize activity. The resulting activity recognition algorithms were evaluated on the test sample. A multi hold-out procedure allowed repeating the operation with 5 different assignments. The analytic conditions were modulated to test (1) different window lengths (1-60 seconds), (2) some selected sensor configurations and (3) different numbers of data features. RESULTS: A window length of 20 s was found to be optimum and therefore used for the rest of the analysis. Using all the sensors and all 167 features, the smart shoes predicted the activities with an average success of 89%. "Running" demonstrated the highest sensitivity (100%). "Walking up a slope" was linked with the lowest performance (63%), with the majority of the false negatives being "walking on a flat surface" and "walking upstairs." Some 2- and 3-sensor configurations were linked with an average success rate of 87%. Reducing the number of features down to 20 does not alter significantly the performance of the algorithm. CONCLUSION: High-performance human behavior recognition using plantar pressure data only is possible. In the future, smart-shoe devices could contribute to the evaluation of daily physical activities. Minimalist configurations integrating only a small number of sensors and computing a reduced number of selected features could maintain a satisfying performance. Future experiments must include a more heterogeneous population.

Development of intraoperative plantar pressure measuring system considering weight bearing axis.

PURPOSE: Surgical reconstructions in three dimensions are needed for treatment of foot and ankle deformities. However, surgical results might be influenced by the skill and experience of doctors which complement the limited information for reconstructions in three dimensions. To solve these, studies were carried out to measure plantar pressure distribution during surgery. Though, it was impossible to accurately measure plantar pressure distribution accurately during operation. Therefore, we proposed an intraoperative plantar pressure measurement (IPPM) device that enables proper navigation in the push direction. METHODS: For this purpose, first, we investigated how the physiological load axis passes through the human body to identify the pushing direction of the pressure sensor of the device toward the patient's foot. In particular, we hypothesized that the physiological load axis passes through the femoral head center and we evaluated this in a measurement experiment with nine healthy subjects. Second, based on these results, we developed the IPPM device that has two force sensors to identify the pushing direction toward the femoral head center and a conductive ink sensor to measure plantar pressure distribution. Finally, we conducted the experiments with nine healthy subjects and two users. RESULTS: From the first experimental results, the physiological load axis was found to pass through the femoral head center in normal standing posture. From the evaluation experiment, there are no significant differences statistically in plantar pressure distributions between the conditions of using IPPM device and without using it for both a medical student and a surgeon. However, in some cases the plantar pressure distribution can be reproduced similarly to that of the standing posture, and also from the evaluation experiment concerning the relation between CoP position and NCC, the NCC tends to increase when the position of the CoP is closer to that at the standing posture. CONCLUSION: The IPPM device has possibility to reproduce the plantar pressure distribution during surgery and prevent the recurrence of surgical complications.

Measuring gait pattern in elderly individuals by using a plantar pressure measurement device.

BACKGROUND: Hip fracture in the elderly is a serious problem, and solutions to prevent falls are needed. OBJECTIVE: This study focused on elucidating data critical to fall prevention by evaluating ambulatory function, and we achieved this by developing a plantar pressure measurement device to determine gait function. METHODS: Our device enables measurement of gait function in the unrestrained state by transmitting wireless data. In this study, we applied the device to field experiments involving 98 subjects (39 healthy individuals, 44 elderly non-fallers, and 15 elderly fallers). Gait features were determined by measuring the pressure values and foot contact patterns used as gait function parameters in previous studies. RESULTS: In particular, decreased peak pressure values were noted at heel strike and toe off during walking in elderly fallers compared with elderly non-fallers. In addition, compared with healthy subjects, elderly fallers also showed extension of the double support phase, and differences in individual gait pattern features were observed between the groups. CONCLUSIONS: Experiments confirmed that our device can be used to obtain the gait features of a diverse group of elderly individuals. Moreover, our device enables objective and quantitative evaluation of gait function and thus may be useful for evaluating gait function in the elderly.