Effective retraining of foot elevation and forward propulsion is a critical aspect of gait rehabilitation therapy after stroke, but valuable feedback to enhance these functions is often absent during home-based training. To enable feedback at home, this study assesses the validity of an inertial measurement unit (IMU) to measure the foot strike angle (FSA), and explores eight different kinematic parameters as potential indicators for forward propulsion. Twelve people with stroke performed walking trials while equipped with five IMUs and markers for optical motion analysis (the gold standard). The validity of the IMU-based FSA was assessed via Bland–Altman analysis, ICC, and the repeatability coefficient. Eight different kinematic parameters were compared to the forward propulsion via Pearson correlation. Analyses were performed on a stride-by-stride level and within-subject level. On a stride-by-stride level, the mean difference between the IMU-based FSA and OMCS-based FSA was 1.4 (95% confidence: −3.0; 5.9) degrees, with ICC = 0.97, and a repeatability coefficient of 5.3 degrees. The mean difference for the within-subject analysis was 1.5 (95% confidence: −1.0; 3.9) degrees, with a mean repeatability coefficient of 3.1 (SD: 2.0) degrees. Pearson’s r value for all the studied parameters with forward propulsion were below 0.75 for the within-subject analysis, while on a stride-by-stride level the foot angle upon terminal contact and maximum foot angular velocity could be indicative for the peak forward propulsion. In conclusion, the FSA can accurately be assessed with an IMU on the foot in people with stroke during regular walking. However, no suitable kinematic indicator for forward propulsion was identified based on foot and shank movement that could be used for feedback in people with stroke.

This approach offers a structured method for organizations to develop and articulate their ethical frameworks, particularly in areas where legal guidance is limited or nonexistent. Problem: This study investigates establishing core values in a legal vacuum, where research, design, or implementation of an invention or innovation is feasible but lacks regulations. We leverage Large Language Models (LLMs) to analyze codes of conduct from 1000 organizations (profit and not-for-profit) to identify core values. Metrics such as accuracy, bias, completeness, consistency, and relevance are used to validate the performance of LLMs in this context. From 493 non-profit organizations and companies on the Fortune 500 list, a total of 8646 core values including variations across 89 sectors were found. Using accuracy, bias, completeness, consistency and relevance as metrics for evaluating result from the LMMs, the number of core values is reduced to 362. The research employs a ten-step decision-making process to guide ethical decision-making when clear rules, laws, or regulations are absent. The framework presents how objectivity can be maintained without losing personal values. This research contributes to understanding how core values are established and applied in the absence of formal regulations.