LightPRA: A Lightweight Temporal Convolutional Network for Automatic Physical Rehabilitation Exercise Assessment.

Research evidence shows that physical rehabilitation exercises prescribed by medical experts can assist in restoring physical function, improving life quality, and promoting independence for physically disabled individuals. In response to the absence of immediate expert feedback on performed actions, developing a Human Action Evaluation (HAE) system emerges as a valuable automated solution, addressing the need for accurate assessment of exercises and guidance during physical rehabilitation. Previous HAE systems developed for the rehabilitation exercises have focused on developing models that utilize skeleton data as input to compute a quality score for each action performed by the patient. However, existing studies have focused on improving scoring performance while often overlooking computational efficiency. In this research, we propose LightPRA (Light Physical Rehabilitation Assessment) system, an innovative architectural solution based on a Temporal Convolutional Network (TCN), which harnesses the capabilities of dilated causal Convolutional Neural Networks (CNNs). This approach efficiently captures complex temporal features and characteristics of the skeleton data with lower computational complexity, making it suitable for real-time feedback provided on resource-constrained devices such as Internet of Things (IoT) devices and Edge computing frameworks. Through empirical analysis performed on the University of Idaho-Physical Rehabilitation Movement Data (UI-PRMD) and KInematic assessment of MOvement for remote monitoring of physical REhabilitation (KIMORE) datasets, our proposed LightPRA model demonstrates superior performance over several state-of-the-art approaches such as Spatial-Temporal Graph Convolutional Network (STGCN) and Long Short-Term Memory (LSTM)-based models in scoring human activity performance, while exhibiting lower computational cost and complexity.

Economic Inequality in Visual Impairment: A Study in Deprived Rural Population of Iran.

PURPOSE: To determine economic inequality in visual impairment (VI) and its determinants in the rural population of Iran. METHODS: In this population-based, cross-sectional study, 3850 individuals, aged 3-93 years were selected from the north and southwest regions of Iran using multi-staged stratified cluster random sampling. The outcome was VI, measured in 20 feet. Economic status was constructed using principal component analysis on home assets. The concentration index (C) was used to determine inequality, and the gap between low and high economic groups was decomposed to explained and unexplained portions using the Oaxaca-Blinder decomposition method. RESULTS: Of the 3850 individuals that were invited, 3314 participated in the study. The data of 3095 participants were finally analyzed. The C was -0.248 (95% confidence interval [CI]: -0.347 - -0.148), indicating a pro-poor inequality (concentration of VI in low economic group). The prevalence (95% CI) of VI was 1.72% (0.92-2.52) in the high economic group and 10.66% (8.84-12.48) in the low economic group with a gap of 8.94% (6.95-10.93) between the two groups. The explained and unexplained portions comprised 67.22% and 32.77% of the gap, respectively. Among the study variables, age (13.98%) and economic status (80.70%) were significant determinants of inequality in the explained portion. The variables of education (coefficient: -4.41; P < 0.001), age (coefficient: 14.09; P < 0.001), living place (coefficient: 6.96; P: 0.006), and economic status (coefficient: -7.37; P < 0.001) had significant effects on inequality in the unexplained portion. CONCLUSIONS: The result showed that VI had a higher concentration in the low economic group, and the major contributor of this inequality was economic status. Therefore, policymakers should formulate appropriate interventions to improve the economic status and alleviate economic inequality.