Gait and balance disorders pose significant risks of injury and negatively impact quality of life. Precise coordination and control of body movement are crucial for bodily-kinesthetic intelligence (BKI). In this work, we propose a fall risk feature mining method (FRFMM) that utilizes foot pressure sensor data and is based on a non-parametric walking model (NPWM). Our approach generates time-spatial features, including raw features, balance features, ratio features, and gait energy expenditure features. We employ three sorted metrics—correlation coefficient in features (CCF), Bayes Factor (BF), and self-information index (SII)—to refine body schema awareness and body image knowledge. This refinement aims to enhance goal-oriented BKI and reduce fall risk. We introduce three threshold rules based on the sorted metrics, enabling the selection of three discriminative fall risk features (the support base of center and two novel ratio features) from 64 generated features derived from plantar pressure sensors. Our framework integrates BKI and FRFMM across four levels: physical observation, feature extraction, decision-making, and multimodal data analysis. Experimental results using a fall risk dataset demonstrate that the FRFMM improves fall risk prediction accuracy by 10% compared to FRFMM with any single metric. Additionally, it outperforms well-known feature selection methods, achieving an accuracy improvement of 7% to 34%. Comparisons with clinical methods further highlight the effectiveness of our approach for analyzing and assessing fall risk.

Gait impairment is a common motor manifestation of Parkinson’s disease (PD), which is also frequently accompanied by other motor abnormalities such as bradykinesia, rigidity, postural instability, and movement asymmetry. These motor impairments are closely associated with reduced mobility and increased fall risk. Although wearable plantar insole sensing provides a promising basis for objective gait assessment, existing studies have mainly focused on conventional time- or frequency-domain descriptors, whereas the nonlinear complexity of gait, laterality-related imbalance, and deviation from normal gait patterns remain insufficiently characterized in an integrated manner. To address this gap, this paper proposes FID-Gait, which is a three-domain fusion framework for PD identification using instrumented insole data. The framework combines automated gait-cycle segmentation with multidomain feature modeling, including a fractal domain for nonlinear gait complexity, a plantar-loading–phase imbalance (PLPI) domain for loading asymmetry and temporal disturbance, and a covariance-adjusted deviation (CAD) domain for measuring deviation from normal gait patterns. Experiments on the PhysioNet Gait in Parkinson’s Disease dataset showed that FID-Gait achieved strong discriminative performance under multiple evaluation protocols. At the gait-cycle level, the selected MLP classifier achieved an accuracy of 99.11% and an F1-score of 99.47%. At the subject level, the selected AdaBoost classifier achieved the highest accuracy of 90.22% and the best F1-score reached 93.02%. Five-fold cross-validation further supported the robustness of the proposed representation, and leave-one-subject-out evaluation provided preliminary evidence of subject-independent generalization. Overall, FID-Gait provides an effective and interpretable framework for PD gait characterization and identification in offline experimental settings.

Multidomain and multimodal identification of the walking gait-cycle states is important for detecting and monitoring locomotion disorders such as Parkinson's disease (PD). We propose a novel multizonal clustering and multi-level thresholding method based on analyzing multizonal plantar load distribution for generating a discrete gait-state time-interval (GSTI) signal to improve PD diagnosis accuracy and the effectiveness of rehabilitation through personalized strategies. Multidomain analysis of the GSTI signal shows a novel coupled I.Baryskievic-H.Li bio-oscillator interpreted as a GSTI-derived signal-level oscillatory signature that may be associated with a central nervous system (CNS)-related locomotor rhythm organization. The bio-oscillator consists of two interconnected oscillations with distinct resonant spectral peaks at specific natural frequencies and phase coupling (nonlinearity) between two frequency components. We propose a multidomain feature level of layered Integrative Body Intelligence (IBI) framework to identify lower and higher-order interactions between gait cycle states. The proposed multimodal data level of IBI involves the proposed acoustic and visual biofeedback based on a novel acoustic harmonic plantar pressure model and a 3D gait state portrait of the GSTI signal used for walking gait monitoring and personalized rehabilitation assessment in PD. Experiments on a publicly available PD plantar-insole dataset show that the Multilayer Perceptron (MLP) model based on the selected multidomain (time-interval, spectral, and bispectral) feature subset achieves classification accuracy (94.44%), and offers a trade-off between model complexity and performance for PD recognition. This result suggests that it is possible to accurately diagnose early-stage PD through merely testing patients' GSTI signal.

Parkinson’s disease (PD) is associated with gait impairment, bilateral asymmetry, and increased gait variability, highlighting the need for objective and interpretable wearable gait assessment. Plantar insole recordings directly capture foot–ground loading, but their use in PD assessment is often limited by global or low-order descriptors that do not fully represent regional loading organization. This study proposes a region-specific information-theoretic framework for PD gait assessment using wearable plantar-pressure insoles. Bilateral plantar insole signals were reorganized into five anatomical regions: heel, rearfoot, midfoot, forefoot, and toe. Self-information index (SII), Shannon entropy (EN), negentropy (NEG), sample entropy (SEN), and Kullback–Leibler divergence (KL) features were extracted to characterize self-information fluctuation, probabilistic uncertainty, non-Gaussian organization, temporal irregularity, and directional distributional discrepancy in plantar-pressure dynamics. The resulting feature representation was evaluated at gait-cycle, walking-recording, and subject-independent levels using conventional classifiers, ablation analysis, subject-balanced cycle aggregation, and an information-theoretic three-dimensional feature-space rule model (ITFS-RM). KNN achieved an accuracy of 0.9668 at the gait-cycle level, and MLP achieved an accuracy of 0.9344 at the walking-recording level. Under stricter subject-independent evaluation, the accuracy was 0.8475, and subject-balanced-cycle aggregation achieved an accuracy of 0.8655. Region-specific analysis and ablation experiments showed spatially heterogeneous HC–PD differences, with the toe region showing the most consistent contribution. SII, KL, and NEG provided stable discriminative contributions, particularly in toe-related and regional-transition features. ITFS-RM provided explicit feature combinations, value ranges, and spatial rule boundaries for interpretable walking-recording level and subject-grouped separation. These results support region-specific information-theoretic analysis as an interpretable representation of plantar-pressure dynamics for PD gait assessment and emphasize the need for subject-wise validation when repeated walking recordings are available.