A Multi-Agent Approach for Self-Diagnosis of a Hydrocephalus Shunting System

The human brain is immersed in cerebrospinal fluid, which protects it from mechanical stresses and helps support its weight through buoyancy. A constant overproduction, blockage or reabsorption difficulty can result in a build-up of fluid in the skull (Hydrocephalus), which can lead to brain damage or even death. Existing treatments rely on passive implantable shunts that drain the excess fluid out of the skull cavity, thus keeping intracranial pressure in equilibrium. Shunt malfunction is one of the most common clinical problems in pediatric neurosurgery. Unfortunately, symptoms of various shunt complications can be very similar thus complicating the diagnosing process. It is proposed to complement the existing implanted valve with an intracranial pressure sensor, flowmeter and transceiver to be able to make a self-diagnosis. By using such method, all current shunt malfunctions should be detected early and the types of these malfunctions would be predicted. Currently, a mechatronic valve with control software is under investigation and it will be a future solution for most of current shunt problems. This paper describes the design of a multi-agent system for self-diagnosis of hydrocephalus shunting system. An intelligent concept for intelligent agents is proposed that would deal with any shunt malfunctions in an independent and efficient way, with different agents cooperating and communicating through message exchange, each agent specialised in specific tasks of the diagnosis process. Six types of agents have been proposed to detect any faults in hierarchical way. This paper proposed one of the most promising methods for the self-diagnosis and monitoring of hydrocephalus shunting system based on a novel multi-agent approach.