Practical Causal Analysis for Biomedical Sensing Based on Human-Machine Collaboration

Abstract

In general, human understand phenomena by considering causalities when they face any problem. In fact, many causal-based applications and solutions have been proposed in keeping with theoretical development. For instance, in industrial domain, Furuta et al. proposed a training support system for plant operation in which trainee's knowledge is represented as two-layered model of task hierarchy and qualitative causality (1998). In medical domain, Thang et al. proposed a medical diagnosis support system based on oriental diagnosis knowledge (2006). In their approach, the causality among some subject’s symptoms and their diagnostic outcome is described by using RBF neural network. Nakajima et al. proposed a generic health management framework named Human Health Management Technology which is applied to not only human being but also manufacturing process, energy consumption management, and so forth (2008b). In addition, Hata et al. suggested a concept named Human Health Care System of Systems which focus on health management, medical diagnosis, and surgical support. In the concept, the human health management technology is discussed from view point of system of systems engineering (2009). Thus, causality acquisition and its utilization among complex systems has a quite important role in optimal management. On another front, from a viewpoint of theoretical development, lots of causal analysis theories have been proposed (Bishop, 2006). Bayesian network describes statistical causality among phenomena observed from certain managed systems, and the statistical causality provides inference and reasoning functions (Pearl, 2001). Graphical model visualizes causality among components in complex systems (Miyagawa, 1991). Fuzzy logic helps intuitive representation of causality which is experts’ tacit knowledge (Zadeh, 1996). As mentioned above, causal analysis theories and their applications and solutions in many domains have been improved for long time. However, causal analysis for designing sensors is not discussed enough yet. Thus, in this chapter, a role of causal analysis in biomedical sensing is discussed. In the rest of this article, in section 2, the importance of human-machine collaboration in causal analysis is described. In the section, problems which we address in this chapter is defined. In section 3, a human-machine collaborative causal analysis is proposed. Then, in section 4 and 5, two kinds of biomedical sensing which employ the human-machine collaborative causal analysis are demonstrated, that is, visceral fat measurement and heart rate monitoring.