- 1、本文档共20页,可阅读全部内容。
- 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
- 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载。
- 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
查看更多
Predicting Depressive Disorders in
the Elderly with Multitask
Recurrent Neural Network
1 1* 2 3 4
Zhongzhi Xu , Qingpeng Zhang , Wentian Li , Mingyang Li , Paul Siu Fai Yip
1. School of Data Science, City University of Hong Kong
2. Wuhan Hospital for Psychotherapy, Wuhan, China
3. IMSE Department, The University of South Florida
4. CSRP, The University of Hong Kong
*qingpeng.zhang@cityu.edu.hk
(Int’l J Medical Informatics, 2019)
Background
Why should we care?
Late-life depression affects over six
million Americans aged ≥65.
Depressive disorder among the elderly
•is one of the most prevalent
healthcare issues.
•is one of the leading causes of
morbidity and mortality.
•is associated with the elevated
risk of other diseases.
•accounts for significant and
growing health care expenditures.
•<10% of depressed elderly patients
eventually receive appropriate
treatment (low recognition rate).
Research Gap
•There is rich research on detecting depressive
disorder for individuals, most of which are based on
•pathophysiologic data
•behavioral logs collected by pervasive computing devices
such as smartphones, smart watches and straps.
•Successful as they are, (temporally) predicting
instead of detecting is still an open question.
Research Gap
•Existing research on depression prediction mainly
focused on the prediction of depression rate and
suicide rate for a population, instead of individuals.
•Useful for high-level decision making
•Cannot directly help individuals.
(Journal of Affective Disorders 2019)
1.5
1.4
1.3
1.2
1.1
1
0.9
0.8
0.7
文档评论(0)