Artificial intelligence technology is becoming more prevalent in health care as a tool to improve practice patterns and patient outcomes. This study assessed ability of a commercialized artificial intelligence (AI) mobile application to identify and improve bodyweight squat form in adult participants when compared to a physical therapist (PT). Participants randomized to AI group (n = 15) performed 3 squat sets: 10 unassisted control squats, 10 squats with performance feedback from AI, and 10 additional unassisted test squats. Participants randomized to PT group (n = 15) also performed 3 identical sets, but instead received performance feedback from PT. AI group intervention did not differ from PT group (log ratio of two odds ratios =  − 0.462, 95% confidence interval (CI) (− 1.394, 0.471), p = 0.332). AI ability to identify a correct squat generated sensitivity 0.840 (95% CI (0.753, 0.901)), specificity 0.276 (95% CI (0.191, 0.382)), PPV 0.549 (95% CI (0.423, 0.669)), NPV 0.623 (95% CI (0.436, 0.780)), and accuracy 0.565 95% CI (0.477, 0.649)). There was no statistically significant association between group allocation and improved squat performance. Current AI had satisfactory ability to identify correct squat form and limited ability to identify incorrect squat form, which reduced diagnostic capabilities.Trial Registration NCT04624594, 12/11/2020, retrospectively registered.

Background: While cognitive decline has been frequently reported in aging research, moderating factors for cognitive changes in healthy aging have been inconclusive. This study evaluated 5 year changes in four cognitive abilities and the potential moderation of age and cognitive reserve (CR) factors on cognitive changes. Methods: Participants included 254 healthy adults initially aged 20 to 80 years. Six tasks estimated each of the four abilities: fluid reasoning, processing speed, memory and vocabulary. The proxies for CR included years of education and IQ. Cognitive changes and moderating factors were examine using multiple indicator latent change score model. Change point analysis pinpointed inflection points after which cognitive changes accelerated. Results: There was significant decline over five years in fluid reasoning, processing speed and memory, with age moderation such that older age was associated with steeper decline. Accelerated decline was observed earlier for reasoning and speed, at ages 58 and 59 years respectively, than for memory, at age 70 years. Vocabulary continued to improve until reaching peak performance at 67 years. For moderation of cognitive changes by CR proxies, while education did not show significant moderation, higher IQ was associated with reduced 5-year decline in reasoning and memory but not processing speed. CR moderation effect was found to be independent of mean cortical thickness. Conclusions: Using a robust statistical model to estimate the latent change in four cognitive abilities over 5 years, the results showed that cognitive reserve rather than brain maintenance is the potential mechanism underlying IQs protective effect on cognitive decline.

No abstract

Emotions are not necessarily universal across different languages and cultures. Mental lexicons of emotions depend strongly on contextual factors, such as language and culture. The Chinese language has unique linguistic properties that are different from other languages. As a main variant of Chinese, Cantonese has some emotional expressions that are only used by Cantonese speakers. Previous work on Chinese emotional vocabularies focused primarily on Mandarin. However, little is known about Cantonese emotion vocabularies. This is important since both language variants might have distinct emotional expressions, despite sharing the same writing system. To explore the structure and organization of Cantonese-label emotion words, we selected 79 highly representative emotion cue words from an ongoing large-scale Cantonese word association study (SWOW-HK). We aimed to identify the categories of these emotion words and non-emotion words that related to emotion concepts. Hierarchical cluster analysis was used to generate word clusters and investigate the underlying emotion dimensions. As the cluster quality was low in hierarchical clustering, we further constructed an emotion graph using a network approach to explore how emotions are organized in the Cantonese mental lexicon. With the support of emotion knowledge, the emotion graph defined more distinct emotion categories. The identified network communities covered basic emotions such as love, happiness, and sadness. Our results demonstrate that mental lexicon graphs constructed from free associations of Cantonese emotion-label words can reveal fine categories of emotions and their relevant concepts.