An advisor for the management of the acute radiation syndrome

Abstract

Many of real problems are such complex that one single agent only is not capable to solve them autonomously. Thus, co-operative problem-solving involving the interaction of human agents and machine agents is required to manage complex tasks, e. g., the acute radiation syndrome. A knowledge-based advisor supporting the physician in charge managing an acute radiation syndrome is presented. The advisor is needed to transfer the knowledge of the present experts into the future. The physician in charge is advised how to decompose the original problem into sub problems and to decide to whom the sub problems to be solved are distributed. The management of the acute radiation syndrome can be decomposed into tasks which are routinely done in every hospital and others which are in general not done in the hospital. Due to the fact that the knowledge-based advisor co-operatively distributes most of the routine tasks to humans and only solves a rare minority by itself the efficiency of the co-operative problem-solving process increases. By applying a cognitive problem-solving model implemented on a blackboard control architecture this behaviour is obtained. The patient card metaphor and the communication via forms, to which a physician is accustomed, facilitates the access for the medical personnel.