ABSTRACT
Noriaki Kuwahara, Naoki Ohboshi, Hiromi Itoh Ozaku, Futoshi Naya, Akinori Abe, and Kiyoshi Kogure ATR Knowledge Science Laboratories 2-2-2, Hikaridai, Seika-cho, Soraku-gun, Kyoto 619-0288 JAPAN
Abstract
In this paper, we propose an algorithm for observing nursing activities with a ubiquitous sensor network and detecting errors in nursing care by comparing observed data with the above model to provide nurses with context-aware support: namely, warnings about the possibility of errors. First, we discuss the usefulness of context-aware technology in a medical field where many temporally and spatially distributed personnel are precisely performing various medical practices. Next we introduce applications of context-aware technology in a medical domain. Then we show our research and development for observing and understanding nursing activities with a ubiquitous sensor network in a hospital and propose a robust error detection algorithm for nursing care procedures. This algorithm warns of mistakes or neglected tasks in nursing procedures using observed data from a ubiquitous sensor network. We assume two types of nursing care errors: “neglected tasks” and “out-of-sequence tasks.” The algorithm focuses on these tasks in potentially fatal errors because such situations demand immediate attention. Its validity has been confirmed
Medical errors annually cause many deaths. In the United States, for example, it is estimated that more people die every year as a result of medical errors than from car accidents, breast cancer, or AIDS [5]. From 13,601 nursing care cases examined, the Japanese Ministry of Health, Labor and Welfare recently defined 10,564 as incidents and accidents. Many medical accidents are caused by neglect or carelessness when confirming safety during medical and nursing care [1], [2], [3], [4]. Such neglect or carelessness often happens when medical personnel are too busy. In Japan, for example, a nurse failed to notice a problem with some artificial respiration equipment because she was too busy dealing with other patients, even though hospital regulations require that such respiratory patients must be checked every fifteen minutes.
