Hiromi Itoh Ozaku

Scenario Violation in Nursing Activities: Nursing Risk Management from the Viewpoint of Chance Discovery

This paper introduces scenarios that from a time series of events under a coherent context of performing nursing risk management. First, we describe general nursing risk management procedures. Then we review our previous nursing accident or incident protection model based on abduction. This paper extends the nursing accident or incident protection model by using the concept of scenario. That is, the model introduces chronological information in knowledge presentation. Then this paper regards a set of nursing activities as a scenario and characterizes a (nursing) accident or incident as a scenario violation. The main purpose of this paper is to present nursing risk management from the viewpoint of scenario violation in the context of chance discovery.

Communication Error Determination System for Multi-layered or Chained Situations

Many medical accidents and incidents occurred due to communication errors. To avoid such incidents, in this paper, we propose a system for determining communication errors. Especially, we propose a model that can be applied to multi-layered or chained situations. First, we provide an overview of communication errors in nursing activities. Then we describe the warp and woof model for nursing task that was proposed by Harada and considers multi-layered or chained situations. Next we describe a system for determining communication errors based on the warp and woof model for nursing task. The system is capable of generating nursing activity diagrams semi-automatically and compiles necessary nursing activities. We also propose a prototype tagging of the nursing corpus for an effective generation of the diagrams. Then we combine the diagram generation with the Kamishibai KeyGraph to determine possible points of the hidden or potential factors of communication errors.

Simulated annealing algorithm for scheduling problem in daily nursing cares

In most Japanese hospitals, different nurses handle the pre-assigned nursing cares in different ways, which directly affect the quality of nursing cares. To our knowledge, there has been less attention on ensuring that nurses provide nursing cares in a timely and accurate fashion. Consequently, in this paper, considering the similarities to the traditional job shop scheduling problems, we will model the daily nursing care scheduling problems and propose an efficient scheduling method based on simulated annealing algorithm. By iteratively local searching based on simulated annealing: (1) permutating the tasks from one nurse to another and (2) permutating the sub tasks handled by a nurse from its original position to another new one, the proposed method is evaluated to be applicable to the nursing care scheduling problems (providing comprehensive, coordinated and cost effective nursing cares to patients).

Simulated Annealing Algorithm for Daily Nursing Care Scheduling Problem

Nursing, with the primary mission to provide quality services to patients, is accompanied by a serial of activities like patients assessment, outcomes identification for patients. In general, there has been less attention on ensuring that nurses provide nursing cares in a timely and accurate fashion. Consequently, in this paper, considering the similarity to the traditional job shop scheduling problems, we will model the daily nursing care scheduling problems and propose an efficient scheduling method based on simulated annealing algorithm. By the comparison of several nursing care plans obtained by the dispatching-rule based methods (which have been recognized to be the implementation of human thoughts), the proposed method is evaluated to be applicable to the nursing care scheduling problems (providing comprehensive, coordinated and cost effective nursing cares to patients).

Scenario Violation As Gaps Between Activity Patterns

In this paper, we propose interactive and visual discovery method of hidden factors for accidents or incidents. Accidents or incidents tend to occur because of very small differences from an ideal situation. From the above viewpoint, we introduce scenario violation model and regard scenario violation as a gap from the proper scenario. Previously, we proposed abduction-based scenario violation determination strategy, but in this paper, we adopt the Kamishibai-KeyGraph to analyze multiple patterns at the same time. Then, we show a gap discovery procedure by using the Kamishibai-KeyGraph.

What should be abducible for abductive nursing risk management

In this paper, we analyze the hypothesis features of dynamic nursing risk management. In general, for risk management, static risk management is adopted. However, we cannot manage novel or rare accidents or incidents with general and static models. It is more important to conduct dynamic risk management where non-general or unfamiliar situations can be dealt with. We, therefore, propose an abductive model that achieves dynamic risk management where new hypothesis sets can be generated. To apply such a model to nursing risk management, we must consider types of newly generated hypotheses because sometimes newly generated hypotheses might cause accidents or incidents. We point out the preferable hypotheses features for nursing risk management.

A Three-layered Model of Nursing Based on Hospital Observation Data

OBJECTIVES Our aim is to investigate causes of medical incidents and construct a knowledge base for preventing malpractice based on monitored data. METHODS To monitor nursing care, we developed an observing system of nursing activities with a ubiquitous sensor network and detecting errors in nursing care. This system is composed of a voice-recording device, mobile sensors and environmental setting type sensors. In cooperation with a hospital in western Japan, we have collected nursing activity data of nurses engaged at a combined ward, including ophthalmology, otolaryngology, and internal medicine for diabetes. After analyzing intravenous drip injection procedure (IVDI procedure) data, we introduce a three-layered model of nursing to understand nursing activities based on observed data. This model consists of three layers, 1) nursing care classification layer: Class, 2) nursing care step layer: Step, and 3) nursing care action layer: Action. This model is designed to take consistency with existing nursing care workflows. RESULTS We implemented a detection system and succeeded in comprehending the workflow of IVDI procedure at the rate of over 95%. This system also can distinguish IVDI workflows performed in parallel by at least two or several nurses. We implemented a picture showing interface of IVDI workflows which can show each patient with a specific color and distinct nurses. CONCLUSIONS Our system succeeded in verification of nursing care steps in IVDI procedure in ratios of more than 95%. Detection errors are due to the sensor system, so it is necessary to use or develop more precise devices.

Cooperation Between Abductive and Inductive Nursing Risk Management

In general risk management, inductive management is usually adopted. If we computationally conduct inductive management, it is vital to collect a considerable number of examples to perform machine learning. In the accident or incident report database home page, we can review various types of accidents or incidents. However, since reports are written by various nurses, the granularity and quality of reports are not sufficient for machine learning. We, therefore, explain the importance of conducting dynamic nursing risk management that can be achieved by abduction, then illustrate cooperation between abductive and inductive types of nursing risk management

Analysis of daily nursing care: a nursing care scheduling algorithm

Nurses are with the primary mission to provide quality services to patients up to 24 hours a day. In general, different nurses (especially between nurse experts and novices) handle the pre-assigned nursing cares in different ways, which directly affect the quality of nursing cares. In this respect, it is necessary to quantitatively elucidate the nursespsila action rules for effectively nursing instruction and quantitative evaluation of staffing levels. In this paper, we will analyze the daily nursing cares to quantitatively illustrate the implicit action rules of nurses on their provision of cares from the viewpoint of scheduling. Unlike the conventional methods mainly based on the interviews/dialogues, we hypothetically model the nursespsila action rules as some traditional dispatching rules, and elucidate the nursespsila action rules by evaluating the similarities of the actual nursing care schedules with planned nursing care ones. By analyzing several actual observed nursing cares, we quantitatively elucidate the action rules of veterans and novices, and furthermore, we compared their differences on the rules.

Relation between abductive and inductive types of nursing risk management

In this paper, we contrast inductive nursing risk management and abductive nursing risk management, point out the importance of the abductive type, and suggest cooperation between them. In general risk management, inductive management is usually adopted. If we computationally conduct inductive management, it is vital to collect a considerable number of examples to perform machine learning. For nursing risk management, risk management experts usually perform manual learning to produce textbooks. In the Accident or Incident Report Database home page, we can review various types of accidents or incidents. However, since reports are written by various nurses, the granularity and quality of reports are not sufficient for machine learning. We, therefore, explain the importance of conducting dynamic nursing risk management that can be achieved by abduction, then illustrate cooperation between abductive and inductive types of nursing risk management.