Verbal Autopsy: First Steps Towards Questionnaire Reduction

被引:0
作者
Cejudo, Ander [1 ]
Trigueros, Owen [1 ]
Perez, Alicia [1 ]
Casillas, Arantza [1 ]
Cobos, Daniel [2 ,3 ]
机构
[1] Univ Basque Country, HiTZ Basque Ctr Language Technol, IXA, Basque Country, Spain
[2] Swiss Trop & Publ Hlth Inst, Basel, Switzerland
[3] Univ Basel, Basel, Switzerland
来源
TEXT, SPEECH, AND DIALOGUE, TSD 2021 | 2021年 / 12848卷
关键词
COUNT;
D O I
10.1007/978-3-030-83527-9_10
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Verbal Autopsy (VA) is the instrument used to collect Causes of Death (CoD) in places in which the access to health services is out of reach. It consists of a questionnaire addressed to the caregiver of the deceased and involves closed questions (CQ) about signs and symptoms prior to the decease. There is a global effort to reduce the number of questions in the questionnaire to the minimum essential information to ascertain a CoD. To this end we took two courses of action. On the one hand, the relation of the responses with respect to the CoD was considered by means of the entropy in a supervised feature subset selection (FSS) approach. On the other hand, we inspected the questions themselves by means of semantic similarity leading to an unsupervised approach based on semantic similarity (SFSS). In an attempt to assess, quantitatively, the impact of reducing the questionnaire, we assessed the use of these FSS approaches on the CoD predictive capability of a classifier. Experimental results showed that unsupervised semantic similarity feature subset selection (SFSS) approach was competitive to identify similar questions. Nevertheless, naturally, supervised FSS based on the entropy of the responses performed better for CoD prediction. To sum up, the necessity of reviewing the VA questionnaire was accompanied with quantitative evidence.
引用
收藏
页码:111 / 123
页数:13
相关论文
共 32 条
[1]   Performance criteria for verbal autopsy-based systems to estimate national causes of death: development and application to the Indian Million Death Study [J].
Aleksandrowicz, Lukasz ;
Malhotra, Varun ;
Dikshit, Rajesh ;
Gupta, Prakash C. ;
Kumar, Rajesh ;
Sheth, Jay ;
Rathi, Suresh Kumar ;
Suraweera, Wilson ;
Miasnikof, Pierre ;
Jotkar, Raju ;
Sinha, Dhirendra ;
Awasthi, Shally ;
Bhatia, Prakash ;
Jha, Prabhat .
BMC MEDICINE, 2014, 12
[2]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[3]   Uncounted causes of death [J].
Byass, Peter .
LANCET, 2016, 387 (10013) :26-27
[4]   Strengthening standardised interpretation of verbal autopsy data: the new InterVA-4 tool [J].
Byass, Peter ;
Chandramohan, Daniel ;
Clark, Samuel J. ;
D'Ambruoso, Lucia ;
Fottrell, Edward ;
Graham, Wendy J. ;
Herbst, Abraham J. ;
Hodgson, Abraham ;
Hounton, Sennen ;
Kahn, Kathleen ;
Krishnan, Anand ;
Leitao, Jordana ;
Odhiambo, Frank ;
Sankoh, Osman A. ;
Tollman, Stephen M. .
GLOBAL HEALTH ACTION, 2012, 5 :1-8
[5]   Class-specific soft voting based multiple extreme learning machines ensemble [J].
Cao, Jingjing ;
Kwong, Sam ;
Wang, Ran ;
Li, Xiaodong ;
Kong, Xiangfei .
NEUROCOMPUTING, 2015, 149 :275-284
[6]   XGBoost: A Scalable Tree Boosting System [J].
Chen, Tianqi ;
Guestrin, Carlos .
KDD'16: PROCEEDINGS OF THE 22ND ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2016, :785-794
[7]  
Clark S.J., 2018, Quantifying the contributions of training data and algorithm logic to the performance of automated cause-assignment algorithms for verbal autopsy
[8]  
Cohen K. B., 2014, BIOMEDICAL NATURAL L, V11
[9]   The case for verbal autopsy in health systems strengthening [J].
D'Ambruoso, Lucia ;
Boerma, Ties ;
Byass, Peter ;
Fottrell, Edward ;
Herbst, Kobus ;
Kallander, Karin ;
Mullan, Zoe .
LANCET GLOBAL HEALTH, 2017, 5 (01) :E20-E21
[10]  
Dalianis H, 2018, CLIN TEXT MINING SEC, P83, DOI [DOI 10.1007/978-3-319-78503-5, 10.1007/978-3-319-78503-5]