Ziheng Wang's research while affiliated with Tohoku University and other places

... On the other hand, lower fitness levels among individuals with type 1 diabetes seem to be explained by other factors than physical activity or sedentary behaviour. Chang et al [73] conducted a scoping review to assess the effects of exercise on CRF in youth with T1D and found four studies, of which only two reported an improvement in CRF. Altered aerobic muscle capacity has been described in children with poor glycemic control compared to those with better control, irrespective of PA levels [22]. ...

... Supervised ML, conversely, relies on labeled data to delineate input-output mappings. 67 Various studies have adopted diverse algorithms: Wang et al. 48 employed linear discriminant analysis, eXtreme Gradient Boosting, and light gradient boosting machine on measures including 10-meter maximum walk speed, grip strength, timed up-and-go test, and 6-minute walk test to determine the post-experimental frailty state of older adults, while Meng et al. 44 utilized the K-nearest neighbors algorithm to predict the frailty status of older participants following an intervention by utilizing pre-intervention features. Moreover, Sun et al. 47 utilized the support vector machines algorithm to predict mini-basketball training program outcomes concerning social communication impairments and restricted and repetitive behaviors in preschool children with autism spectrum disorders. ...

... There is growing evidence that telemedicine in cardio-oncology is effective and well tolerated, even in older patients. [105][106][107] Telemedicine approaches are effective and safe in older cancer patients, with increasing use of artificial intelligence. [108][109][110] However, patients' social skills and social networks, especially in the last decades of life and in independent patients, need to be carefully assessed. ...

... 67 Various studies have adopted diverse algorithms: Wang et al. 48 employed linear discriminant analysis, eXtreme Gradient Boosting, and light gradient boosting machine on measures including 10-meter maximum walk speed, grip strength, timed up-and-go test, and 6-minute walk test to determine the post-experimental frailty state of older adults, while Meng et al. 44 utilized the K-nearest neighbors algorithm to predict the frailty status of older participants following an intervention by utilizing pre-intervention features. Moreover, Sun et al. 47 utilized the support vector machines algorithm to predict mini-basketball training program outcomes concerning social communication impairments and restricted and repetitive behaviors in preschool children with autism spectrum disorders. ...