Physiological rhythms are ubiquitous and essential to our life. They usually interact with one another and also with the outside environment. Disappearance of normal rhythms and emergence of abnormal rhythms are called dynamical diseases. In this article, we will first review the current knowledge on the genesis of physiological rhythms. Then, models of rhythmic interactions among themselves and with external stimuli will be reviewed. Particular emphasis will be placed on the methods that can diagnose abnormal rhythms. Finally, treatment of dynamical diseases will be discussed. It turns out that the models of fractional Brownian motion and fractional Gaussian noise based on dynamical systems have the potential to become biomarkers in differentiating and evaluating normal from abnormal physiological rhythms in dynamical diseases. Meanwhile, in order to explain how acupuncture works, a feasible model of meridians based on communication networks is also included.