In the past few years, the use of wireless sensor nodes for remote health care monitoring has been advocated as an attractive alternative to the traditional hospital-centric health care system from both the economic perspective and the patient comfort viewpoint. The semiconductor industry plays a crucial role in making the changes in the health care system a reality. User acceptance of remote health monitoring systems depends on their comfort level, among other factors. The comfort level directly translates to the form factor, which is ultimately defined by the battery size and system power consumption. This article introduces low-power wireless sensor nodes for biomedical applications that are capable of operating autonomously or on very small batteries. In particular, we take a closer look at component-level power optimizations for the radio and the digital signal processing core as well as the trade-off between radio power consumption and on-node processing. We also provide a system-level model for WSNs that helps in guiding the power optimization process with respect to various trade-offs.