Posters
« Back
Robust Characterization of Inhalation Information Using a Deep Neural Network and a Smartphone
EP28711
Poster Title: Robust Characterization of Inhalation Information Using a Deep Neural Network and a Smartphone
Submitted on 19 Jun 2018
Author(s): Kian Min Lim, Szer Ming Lee, David Harris, Phil Seeney
Affiliations: PA Consulting, Cambridge, UK
This poster was presented at RDD 2018
Poster Views: 890
View poster »


Poster Information
Abstract: The flow rate and inhalation performance of a patient using a dry powder inhaler (DPI) can be inferred by analyzing the sound that is naturally produced by the DPI during the process of inhalation. A novel method is described in Lee et. al [1]. However, this method performs best in a clinical environment in which the background noise is low and well controlled, and does not overwhelm the inhaler’s acoustic signature. In order for acoustic analysis of an inhaler to work robustly in a typical use environment, such as homes, work places, or out in public, where it is subjected to surrounding noise interference, improvement to the extraction and characterization of the DPI acoustic signal is required.

In this paper, the authors present their work of using audio samples of DPI inhalation, recorded in typical home and office environments, to train a deep learning algorithm to enable detection with high confidence that the DPI has been used.
Summary: In this paper, the authors present their work of using audio samples of DPI inhalation, recorded in typical home and office environments, to train a deep learning algorithm to enable detection with high confidence that the DPI has been used.References: REFERENCES
1. Lee SM, Lim KM, Harris D, Seeney P: A novel method using a smartphone to capture and analyse patient inspiratory profiles. In Drug Delivery to the Lungs 2017. The Aerosol Society. Edinburgh, UK: 2017: Poster 35.
2. Zhang H, McLoughlin I, Song Y: Robust sound event recognition using convolutional neural networks. In 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). South Brisbane, QLD: 2015: 559-63.
3. Krizhevsky A, Sutskever I, Hinton GE: ImageNet classification with deep convolutional neural networks. In NIPS’12 Proceedings of the 25th International Conference on Neural Information Processing Systems. Volume 1. Curran Associates: 2012: 197-105.
Report abuse »
Questions
Ask the author a question about this poster.
Ask a Question »

Creative Commons