Explaining Machine Learning Models for Clinical Gait Analysis

Machine Learning (ML) is increasingly used to support decision-making in the healthcare sector. While ML approaches provide promising results with regard to their classification performance, most share a central limitation, their black-box character. This article investigates the usefulness of Explainable Artificial Intelligence (XAI) methods to increase transparency in automated clinical gait classification based on time series. For this purpose, predictions of state-of-the-art classification methods are explained with a XAI method called Layer-wise Relevance Propagation (LRP). Our main contribution is an approach that explains class-specific characteristics learned by ML models that are trained for gait classification. We investigate several gait classification tasks and employ different classification methods, i.e., Convolutional Neural Network, Support Vector Machine, and Multi-layer Perceptron. We propose to evaluate the obtained explanations with two complementary approaches: a statistical analysis of the underlying data using Statistical Parametric Mapping and a qualitative evaluation by two clinical experts. A gait dataset comprising ground reaction force measurements from 132 patients with different lower-body gait disorders and 62 healthy controls is utilized. Our experiments show that explanations obtained by LRP exhibit promising statistical properties concerning inter-class discriminativity and are also in line with clinically relevant biomechanical gait characteristics.

The Path of Film and Television Animation Creation Using Virtual Reality Technology under the Artificial Intelligence

The purpose is to provide a more reliable human-computer interaction (HCI) guarantee for animation works under virtual reality (VR) technology. Inspired by artificial intelligence (AI) technology and based on the convolutional neural network—support vector machine (CNN-SVM), the differences between animation works under VR technology and traditional animation works are analyzed through a comprehensive analysis of VR technology. The CNN-SVM gesture recognition algorithm using the error correction strategy is designed based on HCI recognition. To have better recognition performance, the advantages of depth image and color image are combined, and the collected information is preprocessed including the relations between the times of image training iterations and the accuracy of different methods in the direction of the test set. After experiments, the maximum accuracy of the preprocessed image can reach 0.86 showing the necessity of image preprocessing. The recognition accuracy of the optimized CNN-SVM is compared with other algorithm models. Experiments show that the accuracy of the optimized CNN-SVM has an upward trend compared with the previous CNN-SVM, and the accuracy reaches 0.97. It proves that the designed algorithm can provide good technical support for VR animation, so that VR animation works can interact well with the audience. It is of great significance for the development of VR animation and the improvement of people’s artistic life quality.

Personality, Attitudes, and Behaviors Predicting Perceived Benefit in Online Support Groups (Preprint)

BACKGROUND Online Support Groups (OSGs) are distance-delivered, easily accessible health interventions offering emotional support, informational support, experience-based support, and companionship or network support for patients/caregivers managing chronic mental and physical health conditions. OBJECTIVE This study aimed to examine the relative contribution of extraversion, agreeableness, neuroticism, positive attitudes toward OSGs, and typical past OSG usage patterns in predicting perceived OSG benefit in an OSG for parent caregivers of children with neurodevelopmental disorders. METHODS A mix method longitudinal design was used to collect data from 81 parents across Canada. Attitudes toward OSGs and typical OSG usage patterns were assessed using author-developed surveys administered at baseline, before OSG membership. The personality traits of extraversion, agreeableness, and neuroticism were assessed at baseline using the Ten-Item Personality Inventory (TIPI). Perceived OSG benefit was assessed using an author-developed survey, administered two months after initiation of OSG membership. RESULTS A hierarchical regression analysis found that extraversion was the only variable that significantly predicted perceived OSG benefit. CONCLUSIONS The key suggestions for improving future OSGs were facilitating more in-depth, customized, and interactive content in OSGs.

Factors affecting the adoption extent of the balanced scorecard by Vietnamese small- and medium-sized enterprises

Purpose This study empirically examines factors affecting the extent of balanced scorecard (BSC) adoption in Vietnamese small- and medium-sized enterprises (SMEs) such as top management involvement, an innovative culture, a product innovation strategy, organisational resources, a competitive environment and business network support. This study aims to gain an improved understanding and draw important lessons on BSC adoption for SMEs. Design/methodology/approach Using primary data obtained from a survey of top managers of SMEs that have experienced some forms of BSC adoption, the authors conduct their analysis using exploratory factor analysis and regression analysis methods. Findings The authors find that top management involvement, an innovative culture, organisational resources and business network support are essential factors impacting the extent of BSC adoption in Vietnamese SMEs. Besides confirming literature findings on these variables, the authors identify support of business networks as another important factor affecting the extent of BSC adoption, alongside location and business owners’ experience. However, the impacts of a product innovation strategy and a competitive environment are not significant. Research limitations/implications This study adapts scales previously designed for large enterprises in developed countries to fit into the context of Vietnamese SMEs. Future research can take advantage of this new set of scales and data to obtain further research results. Practical implications This study will serve as guidance for SMEs considering BSC adoption to have a clear vision of what factors are likely to affect BSC adoption, how they affect it and in what direction. Social implications Lessons learned can be extended not only to Vietnamese SMEs that have not yet adopted the BSC but also to firms in other countries with similar economic conditions. Originality/value This study is among pioneering studies on BSC in SMEs and within the context of Vietnam.

Tasks and performance indicators of intelligent neural network support for decisions on managing labor potential of the fishery complex

Реализация стратегии развития рыбохозяйственного комплекса требует использования принципиально новой модели управления его социально-трудовой сферой, основанной на постоянном развитии: 1) человеческого потенциала, 2) производственной среды (процессов и технологий), 3 ) инструменты управления. Наиболее полным и эффективным решением этих задач является концепция управления трудовым потенциалом. Учитывая, что трудовой потенциал - это сложно формализуемый объект, требующий многомодельного подхода, для моделирования процессов управления им целесообразно использовать классические и новые искусственные нейронные сети. В статье представлена ​​многоуровневая структура показателей эффективности интеллектуальной нейросетевой поддержки принятия решений по управлению трудовым потенциалом рыбохозяйственного комплекса и сформулированы одна обобщенная и шестнадцать частных задач, решение которых осуществляется методами нейросетевого моделирования и направлено. при достижении заданных значений показателей эффективности. Implementation of the strategy for the development of the fishery complex requires the use of a fundamentally new model of management of its social and labor sphere, based on continuous development of: 1) human potential, 2) work environment (processes and technology), 3) management tools. The most complete and effective solution to these tasks is the concept of labor potential management. Taking into account that labor potential is a difficult-to-formalize object that requires a multi-model approach, it is advisable to use classical and new artificial neural networks to model the processes of managing it. The article presents a multi-level structure of efficiency indicators of intelligent neural network support for decisions on managing the labor potential of the fishery complex and formulates one generalized and sixteen particular tasks, solution of which is carried out by methods of neural network modeling and aimed at achieving specified values ​​of efficiency indicators.

The Association Between Online Social Network Support and Fear of COVID-19 Among Older Adults

Public health concerns related to the COVID-19 health crisis are particularly salient among older adults. Fear surrounding COVID-19 has also been associated with increased spread, morbidity, and mortality of the disease. Prior to the pandemic, loneliness and social isolation were already a concern for older adults, and the pandemic further constrained how older adults may socially connect with others because of public health safety precautions. Online social networks are a valuable form of support for older adults, and usage of online social networks during the pandemic may have expanded. Thus, the purpose of this study is to examine the association between online social networks and fear of COVID-19 among older adults. A convenience sample (n = 239) of adults 60+ years of age in the U.S. completed a 20-minute, online survey. The independent variable utilized the Lubben Social Network Scale (four items), focusing on online support. The dependent variable was measured by the Fear of COVID-19 scale (eight items). Results of ordinary least squares regression show that increased online social network support was significantly associated with decreased fear of COVID-19 (p < 0.05), while holding constant age, sex, race, marital status, education, whether a respondent lives alone, and self-rated health. Findings highlight the importance of online social networks for older adults during the COVID-19 crisis. Existing online networks which engage older adults should be expanded, and efforts should be made to provide older adults with online forms of social support who may experience barriers or inequities related to accessing technology.