One of the most common complications of stroke patients is swallowing difficulty (dysphagia). In addition, such patients with dysphagia often suffer from a variety of clinical challenges, such as malnutrition, significant weight loss, aspiration pneumonia, and so forth. Most previous studies have described the magnitude and duration of swallowing behavior. However, there is no objective and accurate method to assess whether swallowing abnormalities are accompanied by many of the potential factors that affect the risk of future stroke. To evaluate swallowing behavior more accurately, we designed an Arduino-based portable embedded system for evaluating dysphagia to measure several major muscle groups simultaneously during swallowing. With this system, the changes in the relationship between several major muscle groups are explored as an indicator to effectively reflect the differences in the surface electromyography (sEMG) patterns corresponding to the activities of different muscle groups. The results of this study show that there is a significant difference between healthy people and patients with dysphagia. These indicators of abnormality can be used to effectively assess the difference between the left and right groups of swallowing muscles and the potential risk of stroke in the future.