Vegetarian diets are typically more sustainable than omnivorous ones due to using less environmental resources in the production of food. An important consideration with plant protein and vegetarian diets, however, is whether this would affect athletic performance. To examine this, 70 male and female endurance athletes were compared for maximal oxygen uptake (VO2 max), peak torque when doing leg extensions, and body composition. Vegetarians had higher VO2 max, but peak torque was not significantly different by diet. Omnivores had higher total body mass, lean body mass, and there was a trend for peak torque to be higher.
To investigate whether plant-protein can comparably support development of lean body mass and strength development in conjunction with strength training, 61 healthy young males and females began a 12-week training and protein supplementation study. While previous training studies have shown no differences for lean body mass or strength development when consuming either soy (plant) or whey (animal) protein supplements in very large amounts (>48 grams), when consuming around 15-20 grams, whey has contributed to greater lean body mass accrual, although strength increases remain similar. The present study matched supplements by leucine content instead of by total protein amount since leucine has been shown to be a key stimulator of muscle protein synthesis and is more concentrated in animal protein. There were no significant differences between the whey or soy group for lean body mass or strength development, as assessed using isokinetic dynamometry doing leg extensions and flexions.
Human activity recognition is the task of identifying a person’s movement from sensors in a wearable device, such as a smartphone, smartwatch, or a medical-grade device. A great method for this task is machine learning, which is the study of algorithms that learn and improve on their own with the help of massive amounts of useful data. These classification models can accurately classify activities with the time-series data from accelerometers and gyroscopes. A significant way to improve the accuracy of these machine learning models is preprocessing the data, essentially augmenting data to make the identification of each activity, or class, easier for the model. <br/>On this topic, this paper explains the design of SigNorm, a new web application which lets users conveniently transform time-series data and view the effects of those transformations in a code-free, browser-based user interface. The second and final section explains my take on a human activity recognition problem, which involves comparing a preprocessed dataset to an un-augmented one, and comparing the differences in accuracy using a one-dimensional convolutional neural network to make classifications.