Over the next three years, researchers at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) hope to design a wearable device for monitoring the breathing of patients with chronic diseases such as asthma or bronchitis. Methods from artificial intelligence will be used to automatically classify and evaluate data. The joint project involving researchers at FAU and McGill University Québec (Canada) has received 225,000 euros in funding from the Free State of Bavaria.
Asthma and chronic bronchitis are the most common chronic respiratory diseases worldwide. In Western industrial countries, they affect up to ten percent of the population, with massive knock-on effects for health systems and the economy as a whole. Symptoms tend to fluctuate and can deteriorate rapidly, for example if aggravated by smoking. Symptomatic attacks can be life-threatening and require immediate medical attention.
Researchers at FAU and McGill University Québec hope to design a wearable device that is capable of automatically monitoring respiratory function and reliably recognises changes in symptoms. 'The wearable basically consists of a sensor, a chip and an antenna,' explains Prof. Dr. Andreas Kist, Professor for Artificial Intelligence in Communication Disorders at FAU. 'It is attached to the patient's neck, where it measures the surface movements of the skin and identifies characteristic symptoms such as coughing, sneezing, panting, shortness of breath or a change for the worse in vocal pitch.' The sensor itself is smaller than a finger nail.
AI developed at FAU
Whilst the hardware will be developed and the clinical test carried out in Québec, the group led by Andreas Kist will predominantly be involved in processing data, using methods from artificial intelligence. 'First of all, we will have to convert audio-acoustic data to mechano-acoustic data in order to prepare the AI for real-life signals,' explains Kist. 'We will use existing databases, such as those already available for coughs, and will create our own database using publicly available audio recordings.' Using this wide range of data to optimise deep neural networks should enable the device to classify typical respiratory symptoms independently and reliably.
The wearable is designed to communicate via radio with a mobile app that then analyses the data gained. As Andreas Kist explains, 'in simple terms, the wearable recognises when the patient coughs. The app analyses how often they have coughed over the last few hours, whether this is an indication that the patient's health is deteriorating and whether the patient may in fact be in acute danger.' AI algorithms will also be used during the monitoring process to integrate information from different sources, such as personal data regarding the patient's age, stage of disease and current medication. The wearable is predominantly designed to act as an emergency system, but over the medium term the continuous monitoring is hoped to also cut hospitalisation rates, reduce the numbers of doctor's appointments and therefore save costs in the health system.
More than 200,000 euros in funding
Within the context of the programme 'Artificial Intelligence in Health Research', the Free State of Bavaria has allocated 225,000 euros in funding over the next three years to the project 'AIrway, an AI-powered wearable device for airway health monitoring'. The programme was initiated by the Bavarian Ministry of Science and the Arts in August 2021, together with the foundation for research funding 'Fonds de recherche du Québec' (FRQ). The AIrway project was chosen as one of three projects to be awarded funding from a total of 20 applicants, underlining once again the expertise of FAU not only as a location for innovation and also a hub for artificial intelligence in medicine. The professorship led by Andreas Kist has been established at the Department of Artificial Intelligence in Biomedical Engineering (AIBE). The AIBE was established as part of the High-Tech Agenda Bavaria in late 2019 and takes an interdisciplinary and cross-subject approach at the intersection of medicine and engineering.