日本データベース学会

dbjapanメーリングリストアーカイブ(2019年)

[dbjapan] 看護行動認識チャレンジ


DBJapanの皆様

九工大の井上創造と申します。

ユビキタスコンピューティングのトップ会議Ubicomp2019の併設ワークショップHASCAで、
今年は、モーションキャプチャなどを用いた看護行動認識チャレンも実施します。

ナースコールのケアコム株式会社と共同でデータを集めましたが、
現場ではなく実験室でバランスの良いデータをとりましたので、
比較的精度を出しやすいデータセットになっていると思います。

・5/26にデータダウンロード開始、
・6/30に結果および原稿締め切り

になります。


参加をご検討いただければ幸いです。
-----
Sozo INOUE, Kyutech / Riken AIP    http://sozolab.jp
九州工業大学 大学院生命体工学研究科 / 理化学研究所AIP 井上創造【新アドレス: sozo [at] brain.kyutech.ac.jp
General Chair, Activity and Behavior Computing (ABC2019)  https://abc-research.github.io/
General Chair, IEEE Pervasive Intelligence (PICom2019) http://cyber-science.org/2019/picom/
-----




CALL FOR PARTICIPATION IN NURSE CARE ACTIVITY RECOGNITION CHALLENGE 
As part of the 7th Int Workshop on Human Activity Sensing Corpus and Applications (HASCA)
in conjunction with ACM Ubicomp/ISWC 2019 London, UK
https://hasc-nurse-challenge.github.io/
---------------------------------------------------------------------------------------------------------------------------

* Submission Deadline: June 30, 2019 *
* Workshop Date: September 9 or 10, 2019 *

In an aging population, the demand for nurse workers increases to care for elders. Helping nurse workers make their work more efficient, will help increase elders quality of life, as the nurses can focus their efforts on care activities instead of other activities such as documentation.
Activity Recognition can be used for this goal. If we can recognize what activity a nurse is engaged in, we can partially automate documentation process to reduce time spent on this task, monitor care plan compliance to assure that all care activities have been done for each elder, among others.
The goal of the HASC Nurse Care Activity Recognition Challenge is to recognize 6 nurse activities by means of body movement analysis (accelerometers, motion capture, indoor location and air pressure). Although complex activities can benefit from other types of sensors such as object or environment sensors, in a nursing domain these can be difficult to install. For example, in nursing homes, extra care for not disturbing the residents' environment must be considered and the sudden installation of sensors can disturb them. Moreover, some nurses are working as visiting nurses, visiting each patient at their own homes where we have no control of sensors available or environment. For these reasons, the challenge aims at measuring the feasibility of recognizing nursing activities based on body movement.

Participants should follow the guidelines in the challenge webpage and submit a paper to the HASCA workshop describing the data sources selected, their preprocessing pipeline, the features used and their results. 
All submissions must follow the HASCA format (up to 5 pages).

CHALLENGE PARTICIPATION & SUBMISSION
--------- 
Participants should register to the challenge via IEEE Data Port to access the data and submit their results
https://ieee-dataport.org/competitions/nurse-care-activity-recognition-challenge

The results should be submitted as a csv file with two columns: segment_id and activity_id, where each line represents one segment of the test data. 

PAPER FORMAT & TEMPLATE
------ 
Short technical papers up to 5 pages
All the papers must be formatted as "new single-column" ACM SIGCHI Extended Abstracts format available from the link below.
[LaTeX] https://www.acm.org/publications/proceedings-template
[Word] http://ubicomp.org/ubicomp2019/files/CHI19-EA-sample-file-1.docx

Submissions do not need to be anonymous.
All publications will be peer reviewed together with their contribution to the topic of the workshop.
The accepted papers will be published in the UbiComp 2019 adjunct proceedings, which will be included in the ACM Digital Library.

PAPER SUBMISSION
--------- 
Please submit your papers from https://new.precisionconference.com/submissions
(select SIGCHI/UbiComp 2019/UbiComp 2019 Workshop – HASCA, and push Go button).

IMPORTANT DATES 
--------- 
Submission deadline : June 30, 2019
Camera ready deadline : July 12, 2019
Workshop date: September 9 or 10, 2019