Distributed Aspects of the Artificial Pancreas

Type 1 diabetes (T1D) is a disease caused by the autoimmune destruction of the pancreatic beta cells, permanently disrupting endogenous insulin secretion and requiring patients to inject or infuse insulin to keep blood sugar levels close to normal. Self-treatment of T1D can be regarded as a closed-loop control system with a significant "human-in-the-loop" aspect. Artificial pancreas systems seek to both improve glycemic outcomes and reduce cognitive burden associated with self-treatment by taking advantage of modern glucose monitoring and insulin pump technology. While the systems that have been proposed so far differ in the nature and extent of patient interaction, all such systems involve closing the feedback loop through a portable computational platform with wireless access to both sensors and actuators. In this paper, we review recent developments in the field, focusing on the opportunities and hazards associated with (i) distributed implementation of artificial pancreas algorithms and (ii) shared human-machine control of insulin delivery.