Web-based intelligent medical tutoring systems

Introduction: Internet is nowadays being integrated into all kinds of education and training processes. In health education, medical educators face the difficult requirement of educating future clinicians, while simultaneously limiting medical errors to real patients. At the same time, new pressures in health care have disrupted the traditional model of clinical apprenticeship. This combination of factors has fuelled an increasing interest in applications of technology, including intelligent technologies, to medical training. With the development of WWW technology, Intelligent Medical Tutoring Systems (IMTS) are becoming a mainstream area of research and development. This web-based IMTS are installed and supported in one place being used by thousands of learners all over the world [I]. Although many web-based educational systems have appeared recently, most of them emerged from their predecessor legacy stand-alone systems. Therefore, they not only restrict themselves in functionality, but also fail in taking advantage of the distributed nature of Internet. These systems are usually dedicated to a limited knowledge base and have rigid architecture with little possibility of modifications once released. The purpose of Intelligent Medical Tutoring Systems is to provide rich environments for maximizing learning while minimizing risks to patients. In this work an IMTS for training in Tuberculosis has been developed and integrated in SINCO project [2]. The IMTS provides one-on-one instruction enabling participants to practice their skills by carrying out tasks within a highly interactive learning environment. Moreover, the system assesses each learner's actions and develops a model of their knowledge, skills, and expertise. Based on the learner model, it tailors instructional strategies in terms of both the content and style; providing explanations, hints, examples, demonstrations, and practice problems as needed. The figure 1 shows the architecture proposed. The IMTS is implemented as CBR system. Methods: Case-based reasoning (CBR) [3] is a problem solving paradigm that in many aspects is fundamentally different from other major AI approaches. Instead of relying only on general knowledge of a problem domain or making associations along generalized relationships between problem descriptors and conclusions, CBR is able to use the specific knowledge of previous experiences in solving concrete problem situations (cases). The IMTS is implemented as a multi-agent case-based reasoning system. The relation between CBR systems and Intelligent Medical Tutoring Systems is established by modelling students as cases. The advantage of this approach is that a problem can be easily conceptualised in terms of agents and later on be implemented like a CBR system. The IMTS-CBR updates its base of cases constantly and in consequence, adapting to changes in the environment. Moreover, each one of the phases of the CBR system are automated. In our approach, the agents implement the CBR cycle phases, working like autonomous entities, acting rationally in accordance with its environment perceptions and knowledge status and exchanging information one another providing modularity and cooperation. Results: The Web-IMTS was integrated in SINCO with the objective of teaching Tuberculosis (TB) courses. This disease was chosen because it was identified as a critical public problem in the region where the project was developed. The poor education on TB is one of the fundamental problems of the current TB control programmes. The use of intelligent system allowed training medical students and health workers geographically distant by means of personalized instruction. ITS let the system adapt to the student's preferences offering the teaching strategies according to their profile. Discussion: The development of the IMTS under a modular and flexible architectural approach, allows its use in other medical areas. The potential reuse of existing resources offers significant advantages to the development of knowledge-based educational systems.