Flandrois J, Carret G
Laboratoire de Bactériologie, Faculté de Médecine Lyon-Sud URA 243, Hôtel Dieu, Lyon.
Ann Biol Clin (Paris). 1991;49(3):166-71.
Artificial intelligence is a part of computer science that deals with programs mimicking intelligence of man. Artificial intelligence is now used to check the quality of the determination of antibiotics susceptibility of bacteria. This application is useful because antibiotic susceptibility is subject to biological and technical variation that have to be detected. Three types of reasoning are used either by the biologist or by expert systems: low level quality checking dealing with individual results, microbiological interpretation of the whole set of results and medical interpretation of the results. The use of artificial intelligence in these fields is sustained by the structured nature of the knowledge. Two type of expert systems are already of routine use, either based on production rules (ATB plus EXPERT, bioMerieux, La Balme-les-Grottes, France and SIR, 12A, Montpellier, France), or on object-oriented representation of the knowledge (EXPRIM from our laboratory). The main problem is, as usually in artificial intelligence applications, to transfer human expertise into an adapted knowledge base. The advantage of experts systems over man are their reproducibility of answer and their availability.
人工智能是计算机科学的一部分，涉及模仿人类智能的程序。目前，人工智能被用于检查细菌对抗生素敏感性测定的质量。这种应用很有用，因为抗生素敏感性会受到必须检测到的生物学和技术变异的影响。生物学家或专家系统使用三种类型的推理：处理单个结果的低级别质量检查、对整套结果的微生物学解释以及对结果的医学解释。这些领域中人工智能的应用因知识的结构化性质而得以持续。已经有两种类型的专家系统在常规使用，一种基于生产规则（ATB plus EXPERT，法国生物梅里埃公司，法国拉巴尔姆莱格罗特；以及SIR，法国蒙彼利埃12A），另一种基于知识的面向对象表示（我们实验室的EXPRIM）。与通常在人工智能应用中一样，主要问题是将人类专业知识转化为合适的知识库。专家系统相对于人类的优势在于其答案的可重复性和可用性。
