Wearable sensors are used in gait analysis to obtain spatiotemporal parameters, with gait events serving as critical markers for foot and lower limb movement. Summarizing detection methods is essential, as accurately identifying gait events and phases are key to deriving precise spatiotemporal parameters through wearable technology. However, a clear understanding of how these sensors, particularly angular velocity and acceleration signals within inertial measurement units, individually or collectively, contribute to the detection of gait events and gait phases is lacking. This review aims to summarize the current state of knowledge on the application for both gyroscopes, with particular emphasis on the role of angular velocity signals, and inertial measurement units with both angular velocity and acceleration signals in identifying gait events, gait phases, and calculating gait spatiotemporal parameters. Gyroscopes remain the primary tool for gait events detection, while inertia measurement units enhance reliability and enable spatiotemporal parameter estimation. Rule-based methods are suitable for controlled environments, whereas machine learning offers flexibility to analyze complex gait conditions. In addition, there is a lack of consensus on optimal sensor configurations for clinical applications. Future research should focus on standardizing sensor configurations and developing robust, adaptable detection methodologies suitable for different gait conditions.

A variety of 3D volumetric scanners and smart-device applications are currently being used in podiatry for recording virtual foot data. The accuracy and reliability of these devices vary, resulting in a large variation in the quality of foot scans used for orthotic design. While it is widely believed that a higher quality scanner yields a better scan and thus is expected to produce a more accurate orthotic design, the direct impact of scanning quality on orthotic design has not yet been tested. Therefore, in this study, three commonly used industrial 3D scanners with varying output qualities were used to obtain foot scans of three participants in two weight-bearing conditions. A total of 54 foot scans were obtained, out of which 18 were used to design orthotic insoles using commercial software (FitFoot360). We found variation in the quality of foot scans produced by the different scanners (61.75 ± 2.23% similarity of the foot scans showing a deviation of less than ±1 mm). However, there were no significant differences in the designed foot orthoses within the same weight-bearing condition (83.59 ± 1.97% similarity of the orthotic designs showing a deviation of less than ±1 mm). The medial arch height and heel width differed significantly only when the weight-bearing condition was changed. The findings from this study suggest that the industrial design and production of an orthotic insole using current methods does not depend on the scanning quality of the scanner used but is dependent on the extent of weight bearing.

Accurate plantar pressure distribution is important for biomechanics, gait analysis, rehabilitation, and diabetic foot assessment. However, wearable plantar pressure systems are often limited by sparse sensor layouts due to hardware complexity, power consumption, and user comfort constraints. This study proposes a multimodal deep learning framework for static plantar pressure prediction using plantar geometry information and sparse landmark constraints. A convolutional block attention module U-Net architecture was developed to integrate plantar geometry and sparse landmark modalities through dual-encoder feature fusion with attention refinement. Different network architectures, fusion strategies, and landmark densities were systematically evaluated using a controlled-variable experimental design. Results demonstrated that feature-level fusion consistently outperformed data-level fusion and unimodal configurations across all landmark densities. The proposed model achieved the best performance with a normalized root mean square error of 0.087 using 16 landmarks, and the same model maintained a normalized root mean square error of 0.138 using only two landmarks, indicating promising reconstruction performance even under highly sparse sensing conditions. Marginal contribution and synergy analyses further showed that feature-level fusion more effectively captured complementary interactions between plantar geometry and sparse anatomical guidance, particularly under sparse landmark conditions. These findings suggest that multimodal feature-level fusion provides an effective strategy for sparse-to-dense plantar pressure reconstruction and may support the development of low-cost intelligent insole systems for biomechanical monitoring and clinical applications.