Deep learning approach to estimate foot pressure distribution in walking with application for a cost-effective insole system

Background Foot pressure distribution can be used as a quantitative parameter for evaluating anatomical deformity of the foot and for diagnosing and treating pathological gait, falling, and pressure sores in diabetes. The objective of this study was to propose a deep learning model that could predict pressure distribution of the whole foot based on information obtained from a small number of pressure sensors in an insole. Methods Twenty young and twenty older adults walked a straight pathway at a preferred speed with a Pedar-X system in anti-skid socks. A long short-term memory (LSTM) model was used to predict foot pressure distribution. Pressure values of nine major sensors and the remaining 90 sensors in a Pedar-X system were used as input and output for the model, respectively. The performance of the proposed LSTM structure was compared with that of a traditionally used adaptive neuro-fuzzy interference system (ANFIS). A low-cost insole system consisting of a small number of pressure sensors was fabricated. A gait experiment was additionally performed with five young and five older adults, excluding subjects who were used to construct models. The Pedar-X system placed parallelly on top of the insole prototype developed in this study was in anti-skid socks. Sensor values from a low-cost insole prototype were used as input of the LSTM model. The accuracy of the model was evaluated by applying a leave-one-out cross-validation. Results Correlation coefficient and relative root mean square error (RMSE) of the LSTM model were 0.98 (0.92 ~ 0.99) and 7.9 ± 2.3%, respectively, higher than those of the ANFIS model. Additionally, the usefulness of the proposed LSTM model for fabricating a low-cost insole prototype with a small number of sensors was confirmed, showing a correlation coefficient of 0.63 to 0.97 and a relative RMSE of 12.7 ± 7.4%. Conclusions This model can be used as an algorithm to develop a low-cost portable smart insole system to monitor age-related physiological and anatomical alterations in foot. This model has the potential to evaluate clinical rehabilitation status of patients with pathological gait, falling, and various foot pathologies when more data of patients with various diseases are accumulated for training.

Experimental Study on the Evolution Mechanism of Landslide with Retaining Wall Locked Segment

The failure of locked segment-type slopes is often affected by rainfall, earthquake, and other external loads. Rainfall scours the slope and weakens the mechanical properties of rock-soil mass. At the same time, rainfall infiltrates into cracks of slope rock mass. Under the action of in situ stress, hydraulic fracturing leads to the development and expansion of rock cracks, which increases the risk of slope instability. Under seismic force, the slope will be subjected to large horizontal inertial force, resulting in slope instability. In this paper, a self-developed loading device was used to simulate the external loads such as rainfall and earthquake, and the model tests are carried out to study the evolution mechanism of landslide with retaining wall locked segment. Three-dimensional laser scanner, microearth pressure sensors, and high-definition camera are applied for the high-precision monitoring of slope shape, deformation, and stress. Test results show that the retaining wall locked segment has an important control effect on landslide stability. The characteristics of deformation evolution and stress response of landslide with retaining wall locked segment are analyzed and studied by changing the slope shape, earth pressure, and the displacement cloud map. The evolutionary process of landslide with retaining wall locked segment is summarized. Experimental results reveal that as the landslide with retaining wall locked segment is at failure, the upper part of the landslide thrusts and slides and the retaining wall produces a locking effect; the middle part extrudes and uplifts, which is accompanied with shallow sliding; and compression-shear fracture of the locked segment leads to the landslide failure.

On the quality of RS41 radiosonde descent data

Radiosonde descent profiles have been available from tens of stations for several years now – mainly from Vaisala RS41 radiosondes. They have been compared with the ascent profiles, with ECMWF short-range forecasts and with co-located radio occultation retrievals. Over this time, our understanding of the data has grown, and the comparison has also shed some light on radiosonde ascent data. The fall rate is very variable and is an important factor, with high fall rates being associated with temperature biases, especially at higher altitudes. Ascent winds are affected by pendulum motion; on average, descent winds are less affected by pendulum motion and are smoother. It is plausible that the true wind variability in the vertical lies between that shown by ascent and descent profiles. This discrepancy indicates the need for reference wind measurements. With current processing, the best results are for radiosondes with parachutes and pressure sensors. Some of the wind, temperature and humidity data are now assimilated in the ECMWF forecast system.

Sensitivity of Piezoresistive Pressure Sensors to Acceleration

The article presents the negative aspects of the influence of static and dynamic acceleration on the accuracy of pressure measurement for a selected type of transmitter. The influence of static accelerations from catalog notes was shown and compared with the tests results for a few selected sensors. The results of research on the influence of dynamic acceleration for various types of its variability for selected converters are presented. Moreover, a method of measurement patented by the authors that uses a complex transducer is shown. The method allows for more accurate measurements on moving objects. The tests were performed based on the proposed method. The obtained results of the influence of acceleration on the classical sensor as well as the construction using the proposed method are shown. The paper presents approximate pressure measurement errors resulting from the influence of acceleration. For example, errors in measuring the speed of an airplane may occur without the proposed method. The last part of the article presents a unique design dedicated to a multi-point pressure measurement system, which uses the presented method of eliminating the influence of accelerations on the pressure measurement.

Flexible Capacitive Pressure Sensors with Micro-Patterned Porous Dielectric Layer for Wearable Electronics

Highly sensitive soft sensors play key roles in flexible electronics, which therefore have attracted much attention in recent years. Herein, we report a flexible capacitive pressure sensor with high sensitivity by using engineered micro-patterned porous polydimethylsiloxane (PDMS) dielectric layer through an environmental-friendly fabrication procedure. The porous structure is formed by evaporation of emulsified water droplets during PDMS curing process, while the micro-patterned structure is obtained via molding on sandpaper. Impressively, this structure renders the capacitive sensor with a high sensitivity up to 143.5 MPa-1 at the pressure range of 0.068~150 kPa and excellent anti-fatigue performance over 20,000 cycles. Meanwhile, the sensor can distinguish different motions of the same person or different people doing the same action. Our work illustrates the promising application prospects of this flexible pressure sensor for the security field or human motion monitoring area.

Pressure Sensor Placement for Leak Localization in Water Distribution Networks Using Information Theory

This paper presents a method for optimal pressure sensor placement in water distribution networks using information theory. The criterion for selecting the network nodes where to place the pressure sensors was that they provide the most useful information for locating leaks in the network. Considering that the node pressures measured by the sensors can be correlated (mutual information), a subset of sensor nodes in the network was chosen. The relevance of information was maximized, and information redundancy was minimized simultaneously. The selection of the nodes where to place the sensors was performed on datasets of pressure changes caused by multiple leak scenarios, which were synthetically generated by simulation using the EPANET software application. In order to select the optimal subset of nodes, the candidate nodes were ranked using a heuristic algorithm with quadratic computational cost, which made it time-efficient compared to other sensor placement algorithms. The sensor placement algorithm was implemented in MATLAB and tested on the Hanoi network. It was verified by exhaustive analysis that the selected nodes were the best combination to place the sensors and detect leaks.