Fall Detection With UWB Radars and CNN-LSTM Architecture

• Published in: IEEE journal of biomedical and health informatics Vol. 25; no. 4; pp. 1273 - 1283
• Main Authors: Maitre, Julien, Bouchard, Kevin, Gaboury, Sebastien
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.04.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Apartments; Artificial neural networks; Brain injury; Burns; Cameras; Classification; CNN-LSTM; Detection; End of life; Fall; Fall detection; Falls; Feature extraction; Hemorrhage; Informatics; Injuries; Leave-one-subject-out; Long short-term memory; Model testing; Neural networks; Pain; Short term; Three-dimensional displays; Trauma; Ultra wideband radar; Ultrawideband
• DOI: 10.1109/JBHI.2020.3027967
• ISSN: 2168-2194; 2168-2208

Fall detection is a major challenge for researchers. Indeed, a fall can cause injuries such as femoral neck fracture, brain hemorrhage, or skin burns, leading to significant pain. However, in some cases, trauma caused by an undetected fall can get worse with the time and conducts to painful end of life or even death. One solution is to detect falls efficiently to alert somebody (e.g., nurses) as quickly as possible. To respond to this need, we propose to detect falls in a real apartment of 40 square meters by exploiting three ultra-wideband radars and a deep neural network model. The deep neural network is composed of a convolutional neural network stacked with a long-short term memory network and a fully connected neural network to identify falls. In other words, the problem addressed in this paper is a binary classification attempting to differentiate fall and non-fall events. As it can be noticed in real cases, the falls can have different forms. Hence, the data to train and test the classification model have been generated with falls (four types) simulated by 10 participants in three locations in the apartment. Finally, the train and test stages have been achieved according to three strategies, including the leave-one-subject-out method. This latter method allows for obtaining the performances of the proposed system in a generalization context. The results are very promising since we reach almost 90% of accuracy.