<?xml version="1.0" standalone="yes"?>
<Paper uid="I05-6005">
  <Title>Building Dialogue Corpora for Nursing Activity Analysis</Title>
  <Section position="3" start_page="41" end_page="43" type="metho">
    <SectionTitle>
2 Corpora Collection
2.1 E-nightingale Project
</SectionTitle>
    <Paragraph position="0"> In medical sites, higher levels of knowledge and wider experience are needed for using sophisticated medical devices and, in turn, for accommodating a diverse aging society. It is said that insufficient knowledge and experience both influence the performance of medical professionals and causes malpractice. On the other hand, by promoting the use of wearable sensors and environment sensors, it is possible to collect a huge amount of data on actions in the real world. Accordingly, some attempts have been made to utilize knowledge obtained by analyzing daily actions in the course of developing sensors for education and accident prevention.</Paragraph>
    <Paragraph position="1"> If important information could be obtained by analyzing nursing activities automatically recorded by wearable computers, it would be possible to automatically and more correctly make nursing reports. Furthermore, if the nursing reports could be objectively analyzed, the cost of investigating nursing practices could be kept down and more effective survey research on nursing activities could be conducted. Furthermore, it would be possible to protect nurses from malpractice claims by distinguishing between nursing activities during normal situations and during emergencies. null To exploit this potential, we launched the &amp;quot;Enightingale project&amp;quot; to share and construct knowledge obtained by analyzing the daily actions and experiences of nurses. This project aims to develop the following technologies:  1. To observe and understand daily nursing activities.</Paragraph>
    <Paragraph position="2"> 2. To build a knowledge base through understanding these activities.</Paragraph>
    <Paragraph position="3"> 3. To utilized the necessary knowledge when  it's actually needed.</Paragraph>
    <Section position="1" start_page="41" end_page="43" type="sub_section">
      <SectionTitle>
2.2 Obtained Voice Data of Nurses'
Activities
</SectionTitle>
      <Paragraph position="0"> We have recorded various action data by wearable computers designed to analyze nursing activities in hospitals(N. Kuwahara, et al., 2003; N. Kuwahara, et al., 2004). Wearable computers can collect various types of data such as location, passometer results, and posture inclination. In our previous research, we used passometer, location, and cue words to grasp nursing activities. We could understand the nursing activities of a particular nurse but could not understand the relations and information flow among staff in a hospital.</Paragraph>
      <Paragraph position="1"> Dialogues between nurses are very important in the performance of nursing activities and maintaining staff relations. Dialogues between nurses  and other people, such as patients, doctors, and patients' family, are also important for obtaining a variety of conversation behaviors more efficiently.</Paragraph>
      <Paragraph position="2"> We believe that we can understand not only nursing activities but also conversation mechanisms by developing and analysing corpora of nurses' dialogues.</Paragraph>
      <Paragraph position="3"> Therefore, we have conducted experiments to collect voice data of nurses talking during their activities. Our reasons and aims for focusing on voice data are as follows: * To more effectively make lengthy recordings than sensors with consumer model IC  In the following section, features of voice data collected in experiments are explained in detail.  During nurses' shift changes, they hold clinical meetings to discuss patient information, in the process modifying and confirming this information. Furthermore, if problems occur in their work area, they hold brief conferences to solve the problems by discussing their experiences.</Paragraph>
      <Paragraph position="4"> We have recorded nurses dialogues in clinical meetings and brief conferences by using special devices. At the same time, we have obtained such information as who participated in the clinical meetings and brief conferences.</Paragraph>
      <Paragraph position="5"> In the hospital where we carried out our experiments, clinical meetings are held in the morning and the evening time for about 20 minutes to one hour. The brief conferences are held at lunch time for about 30 minutes. Collected data were 80 trials for a one-week experiment. The entire recording time was about 20 hours. We made transcriptions of the collected data. Some sample data are shown in table 11 The transcription was made by four staff members, including an experienced specialist in making transcriptions, a nurse who had worked in hospitals for three years or longer, a pharmacist, and a part-time employee, and it was assumed that the latter three had no experience working in the field of transcription.</Paragraph>
      <Paragraph position="6">  To understand nursing activities, a one-day schedule of nursing activities should be recorded and analyzed. However, it is difficult to transcribe all such recorded data. Therefore, we recorded the voice data of nursing activities along with clues annotated by nurses using the special devices. null To obtain voice data in hospitals, we developed an IC-recorder, a microphone with an event button, and an intermediate control box with a buzzer. The event button is used for explicit voice annotation when nurses start or complete a task. W hen the button is pushed, the buzzer sounds once and its sound is recorded, and then nurses record their tasks at the moment by speaking short sentences. The buzzer is also set to sound periodically (every 10 minutes) to prompt nurses to make voice input about their ongoing tasks. Simple signal processing can extract and classify event-driven and periodic voice records, as well as nurse call rings.</Paragraph>
      <Paragraph position="7"> Event-driven voice data recording is very use- null ful for accurately recording nursing activities. In addition, this approach can record nursing activities at the desired time. As a result, time information can be easily observed, and we can collect objective data on nursing activities. Therefore, our approach improves the efficiency of research on time utilization.</Paragraph>
      <Paragraph position="8"> In one department of a hospital, we have conducted experiments on collecting data of nursing work through voice annotations. The nurses work in three shifts, assigned to primary patients of each ward. All nurses were given instructions on the usage of our devices. The entire recording time was about 500 hours. Collected data were gathered from 39 trials for a one-week experiment involving 15 nurses using our devices. We also made transcriptions of the collected data. Some sample transcriptions are shown in Table.2. The transcription was also made by four staff members as mentioned above in 2.2.1.</Paragraph>
    </Section>
  </Section>
  <Section position="4" start_page="43" end_page="46" type="metho">
    <SectionTitle>
3 Features of the Corpora
</SectionTitle>
    <Paragraph position="0"> As discussed in the previous sections, we focus on dialogue data such as natural conversations collected in daily life or work. We can observe the following features in the corpora.</Paragraph>
    <Paragraph position="1">  * People in the conversations come from various levels of social status.</Paragraph>
    <Paragraph position="2"> * Conversations are conducted in various places or situations.</Paragraph>
    <Paragraph position="3"> * There are certain relationships or dependencies between the current conversation and the previous conversations.</Paragraph>
    <Paragraph position="4"> * Conversations on similar topics are repeated.  For example, nurses engage in conversation not only with other nurses but also with patients, families of patients, and doctors. Naturally, they behave according to their social role. As a result, features of conversation will be different. Examples of such differences are shown as follows:  * Conversations between doctors and nurses include formal and many medical terms.</Paragraph>
    <Paragraph position="5"> * Among nurses at meetings, formal and many abbreviated expressions of medical terms.</Paragraph>
    <Paragraph position="6"> * Among nurses, informal and many abbreviated expressions of medical terms.</Paragraph>
    <Paragraph position="7"> * With patients, families of patients and nurses, formal and many words expressing  medical terms in simple words.</Paragraph>
    <Paragraph position="8"> Nurses exchange and share information on their patients depending on their work situations. The conversations are made in nurse stations, beside a patient's bed, in a clinic, and so on. Sometimes conversation begins from necessity and other times it begin when nurses happen to encounter.</Paragraph>
    <Paragraph position="9"> Information on patients should smoothly be transferred from one shift nurse to the next shift nurse. As a result, conversations dealing with the same or similar matter should have certain relationships and dependencies. For example, topics related to the same patient and the same operation patterns of a certain disease repeatedly appeared in our corpora. For example, &amp;quot;to test blood sugar levels&amp;quot;, in Table 2, appeared whenever the patients finished a meal.</Paragraph>
    <Section position="1" start_page="44" end_page="44" type="sub_section">
      <SectionTitle>
Time Utterance
</SectionTitle>
      <Paragraph position="0"> 11:28:11 I'm going to prepare a set of drip infusion for Abe-san.</Paragraph>
      <Paragraph position="1"> 11:32:01 I've finished preparing the drip for Abe-san.</Paragraph>
      <Paragraph position="2"> 11:32:04 I'm going to test blood sugar levels of Naya-san and Kuwahara-san. 11:33:48 I've finished testing blood sugar levels of Naya-san and Kuwahara-san. 4 Applicability of the Corpora as</Paragraph>
    </Section>
    <Section position="2" start_page="44" end_page="44" type="sub_section">
      <SectionTitle>
Spoken Language Corpora
</SectionTitle>
      <Paragraph position="0"> In this section, we discuss the applicability of the obtained corpora as spoken language corpora and a strategy to build a set of ontologies from them.</Paragraph>
      <Paragraph position="1"> We have already manually generated around 30 hours of corpora for event-driven data and 4 hours of dialogue corpora at clinical meetings.</Paragraph>
      <Paragraph position="2"> To understand nursing activities from event-driven data, we used tags to understand situations. We assumed that key information to evaluate a situation included time, place, medication, disease and the person's name. Therefore, we extracted the person's name, medication and disease from our corpora by using a named entity recognition tool.</Paragraph>
      <Paragraph position="3"> Named entity recognition is useful for extracting such information as the person's name, company name, date, and place name. Named entity recognition tools have been developed in research on information extraction, machine translation, and so on. There are two types of named entity recognition: rule-based recognition and statistics-based recognition. We believe that some connections with personal name, for example &amp;quot; ^,^,X(Mr/Ms)&amp;quot;, are determined at some level, and our target is data that can be exploited to make hand-crafted rules such as medical charts, attendance sheets and so on. Therefore, we studied how to extract personal name, medication and disease using a rule-based named entity recognition tool. NExT is a well-known tool in Japan for rule-based named entity recognition. We used the NExT version 0.82 (NExT a Named Entity Extraction Tool, 2002), which can be download from the Mie University web site.</Paragraph>
      <Paragraph position="4"> The NExT tool utilizes Chasen version 2.3.3 for morphological analysis and ipadic version 2.7.0 as a dictionary(Morphological Analysis System ChaSen, 2003). The dictionary includes many medicine names, diseases, personal names, and place names.</Paragraph>
    </Section>
    <Section position="3" start_page="44" end_page="45" type="sub_section">
      <SectionTitle>
4.1 Tags in the Corpora
</SectionTitle>
      <Paragraph position="0"> Next, we extracted personal names, medication, and diseases from the corpora and tagged the corpora to better understand situations. For example, the W HO tag means a speaker, in this case a nurse as a subject using our special device, the W HOM tag means a hearer, the PT tag means the person or patient name mentioned, and the PLACE tag means the place where the communication is conducted. The TIME tag indicates the absolute time calculated from the IC recorder's elaped time and the experiment's start time. The MED tag means medication, and the DIS tag indicates the disease name. Table 3 shows a sample of the corpora.</Paragraph>
      <Paragraph position="1"> We manually extracted the person name for W HO, W HOM, and PT from the transcription with medical charts. Person names were also extracted by a named entity recognition tool to check the NExT accuracy. Also, we evaluated PLACE by ambience sounds of voice data and schedule of nursing activities. In addition, we manually extracted medication and disease for MED and DIS.</Paragraph>
      <Paragraph position="2"> It should be noted that we are analyzing words specialized to nursing activities in the primitive corpora. Nursing terms include daily used terms for expressing work that supports patients' daily  levels ofPTNaya-san/PTandPTKuwahara-san/PT.</Paragraph>
      <Paragraph position="3"> lives as well as medical terms specialized for the particular medical situation. Furthermore, some medicine and disease names are different from the expressions used in from dictionaries, since our corpora include colloquial expressions. Moreover, it is difficult to tag some expressions that have the same meaning.</Paragraph>
    </Section>
    <Section position="4" start_page="45" end_page="46" type="sub_section">
      <SectionTitle>
4.2 Nursing Terms Featured in the Corpora
</SectionTitle>
      <Paragraph position="0"> In this section, we describe some expressions that express controversial features for tagging.</Paragraph>
      <Paragraph position="1"> For instance, in general &amp;quot;dU(sengan)&amp;quot; means &amp;quot; %(washing face),&amp;quot; but if it is narrated before surgery in ophthalmology, it means &amp;quot; (washing eye).&amp;quot; &amp;quot;qb(jiritsu-suru)&amp;quot; means &amp;quot;standing walk&amp;quot; if it is narrated in a conversation among nurses, but it means &amp;quot;to earn one's living by oneself&amp;quot; in general, as when it is narrated in a conversation between a nurse and a patient's family. Thus some words have multiple meanings according to the situation. As a result, sometimes nurses, especially novice nurses, experience misunderstanding in their communications.</Paragraph>
      <Paragraph position="2"> Another example is &amp;quot;O(horyu)&amp;quot;, which means &amp;quot;suspend giving medicine&amp;quot; or &amp;quot;keep a sting stung&amp;quot;. Therefore, we think we will be able to gather many words that have multiple meanings when we check the collected corpora. This will allow us to build or extend the contents of a dictionary of multiple-meaning expressions.</Paragraph>
      <Paragraph position="3"> In our studies. we also observed the following cases:  * multiple expressions - &amp;quot;(seki)&amp;quot; and &amp;quot;Igaisou&amp;quot; mean tussis and coughing - &amp;quot;(nyou)&amp;quot; and &amp;quot;(harun)&amp;quot;, and &amp;quot; S`l\oshikko)&amp;quot;ymean urine and pee - &amp;quot;v(tounyou)&amp;quot; and &amp;quot;(diabe)&amp;quot; and &amp;quot;DM&amp;quot;ymean sugar diabetes (diabetic mellitus) These types of expressions are used to hide the real meaning from patients, and they're sometimes used because the expression is currently fashionable.</Paragraph>
      <Paragraph position="4"> * abbreviated expressions - &amp;quot;an&amp;quot;ymeans ampule - &amp;quot;miri)&amp;quot; means milliliter or milligram null - &amp;quot; y(bun-ni)&amp;quot; means half or two times - &amp;quot;(nobo-aaru)&amp;quot; means No null volin R, which is a medication for diabetes null These types of expressions seem to be used for quick communication. In particular, medications are sometime shortened into shorter words, such as Novolin R into R only.</Paragraph>
      <Paragraph position="5"> These expressions are relatively frequently used and sometimes seem to cause nursing accidents or incidents due to miscommunication. We think we will be able to obtain many concepts that have multiple expressions and abbre- null viated expressions by checking the collected corpora. Then we could build or extend contents of a multiple-expression concept dictionary and an abbreviated-expression dictionary.</Paragraph>
      <Paragraph position="6"> To develop a support system for nursing activities, we should standardize technical nursing terms. However, there are many types of terms used in typical nurses' dialogues. For example,  Furthermore, there are many new medicines used in hospitals. It is difficult to uniformly manage these new words. The standardization of nursing terms thus has many problems.</Paragraph>
      <Paragraph position="7"> We think that even ambiguous words can be understood if we recognize nurses' working situations by using the information of many types of sensors. This would require using a position sensor to identify nurses' working locations as well as who is participating in conversations. We could supplement missing information and complete or correct ambiguous information by referring to such information from sensors.</Paragraph>
    </Section>
    <Section position="5" start_page="46" end_page="46" type="sub_section">
      <SectionTitle>
4.3 Feature as Spoken Language Corpora
</SectionTitle>
      <Paragraph position="0"> A chronological relationship or dependency map among nurses can be accurately obtained by referring to nursing records and medical charts. Conventional dialogue corpora can only offer one-to-one conversations, permitting only a simple analysis. In addition, it is easy to make a wrong analysis. null On the other hand, if we analyze our corpora with a chronological relationship or dependency among nurses, we can discover true features of conversation phenomena. For instance, person A exchanges information with person B, but person B misunderstands person A's information. Person B conveys the information to person C. Then C realizes B's misunderstanding and informs B so that B can correct the information. Thus we can obtain communication patterns between more than two persons who do not know each other's situation.</Paragraph>
      <Paragraph position="1"> To utilize the dictionaries obtained from our corpora and sensor information, we can build a set of ontologies for conversations between multiple persons. In building a set of ontologies, the mechanism of conversations can be clarified, and a method for finding points of mistakes given ambiguous expressions can be examined.</Paragraph>
      <Paragraph position="2"> In this paper, we focus on nursing vocabularies and nursing activities. Of course they are slightly different from general terms and general situations, but these corpora and dictionary-building techniques can be applied to general terms and situations betweenyspoken language and written language.</Paragraph>
      <Paragraph position="3"> In finishing the transcription of the clinical meetings and event-driven data, we are building the corpora. Here we note that it is difficult to tag MED and DIS because there are multiple expressions and abbreviated expressions. We need to develop a way to designate standard expressions from the multiple and abbreviated expressions.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>