Over 40% of asthma patients in the US have uncontrolled asthma, accounting for three-quarters of the $56 billion annual cost of asthma. Some of the reasons for poor asthma control are: patients and providers overestimate the level of asthma control, which means that the treatment provided is less aggressive than required; providers overestimate patients’ level of adherence, failing to address any misunderstandings, concerns or preferences of the patient; patients underestimate the severity of their asthma symptoms.
Our solution is a wearable device that automatically detects and tracks cough, wheezing severity, lung function and inhaler use. The device provides objective information about the patient’s level of asthma control and level of adherence to the asthma plan.
Technical areas of interest
- Embedded systems design
- Edge AI
- Machine learning
- Wearable devices
- An Optimized Recurrent Unit For Ultra-Low-Power Keyword Spotting, J. Amoh, K. Odame, in Journal Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technology, vol. 3, no. 2, p. 36, 2019.
- Deep Neural Networks For Identifying Cough Sounds J. Amoh, K. Odame," IEEE Transactions on Biomedical Circuits and Systems, vol. 10, no. 5, p. 1003-1011, 2016.
- DeepCough- A Deep Convolutional Neural Network in A Wearable Cough Detection System, J. Amoh, K. Odame, Proc. BioCAS '15 IEEE Biomedical Circuits and Systems, 2015.
- Technologies for Developing Ambulatory Cough Monitoring Devices, J. Amoh, K. Odame, Critical Reviews in Biomedical Engineering vol. 41, no. 6, p. 457–468, 2014.
- Mobile System For Self-Monitoring Of Asthma, T. Ou, J. Amoh, M. Khurana, V. Yadav, and K. Odame, Proc. EMBC '14 IEEE International Conference of the Engineering in Medicine and Biology Society, 2014.