A comparison of public datasets for acceleration-based fall detection. Issue 9 (September 2015)

• Record Type: Journal Article
• Authors: Igual, Raul; Medrano, Carlos; Plaza, Inmaculada
• Is Part Of: Medical engineering & physics. Volume 37:Issue 9(2015:Sep.)
• Page Start: 870
• Page End: 878
• Publication Date: 2015-09
• Subjects: Biomedical engineering -- Periodicals; Biomedical Engineering -- Periodicals; Physics -- Periodicals; Génie biomédical -- Périodiques; Biomedical engineering; Electronic journals; Periodicals; 610.28
• DOI: 10.1016/j.medengphy.2015.06.009
• Languages: English
• ISSNs: 1350-4533

Abstract

<abstract abstract-type="author-highlights" id="abs0001"> <title id="cesectitle0001">Highlights</title> <sec> <p id="spara0001"> <list id="celist0001"> <list-item id="celistitem0001"> <label></label> <p id="para0001">This study compares several accelerometer-based datasets for fall detection.</p> </list-item> <list-item id="celistitem0002"> <label></label> <p id="para0002">The dataset used to test a detector has some influence on its performance.</p> </list-item> <list-item id="celistitem0003"> <label></label> <p id="para0003">The performance decreases if the training and validation datasets are different.</p> </list-item> <list-item id="celistitem0004"> <label></label> <p id="para0004">Large differences are found in the generalization capability of different datasets.</p> </list-item> <list-item id="celistitem0005"> <label></label> <p id="para0005">The techniques considered are not affected by the sampling frequency or the range of the accelerometer.</p> </list-item> </list> </p> </sec> </abstract>