PRAXIS: Towards automatic cognitive assessment using gesture recognition

作者：

Highlights：

• A computer-assisted cognitive assessment method based on Praxis test is proposed.

• Four methods are developed to evaluate dynamic and static gestures in Praxis test.

• A challenging RGB-D dataset is collected consisting of 60 subjects and 29 gestures.

• The deep learning based method outperformed the other approaches.

• The obtained results ensure that automatic cognitive assessment is feasible.

摘要

•A computer-assisted cognitive assessment method based on Praxis test is proposed.•Four methods are developed to evaluate dynamic and static gestures in Praxis test.•A challenging RGB-D dataset is collected consisting of 60 subjects and 29 gestures.•The deep learning based method outperformed the other approaches.•The obtained results ensure that automatic cognitive assessment is feasible.

论文关键词：Human computer interaction,Computer assisted diagnosis,Cybercare industry applications,Medical services,Patient monitoring,Pattern recognition

论文评审过程：Received 28 September 2017, Revised 29 January 2018, Accepted 30 March 2018, Available online 31 March 2018, Version of Record 10 April 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.03.063