Vera Goebel1, Sigurd Aarrestad2, Harriet Akre2, Mohan Kankanhalli3, Stein Kristiansen1,
Thomas Plagemann1, Karl Øyri4
1University of Oslo, Department of Informatics (UiO-Ifi)
2University of Oslo, Institute of Clinical Medicine (UiO-ICM)
3National University of Singapore, Department of Computer Science (NUS)
4Oslo University Hospital, Intervention Centre (OUS)
Diagnosing OSA is usually done by hospitalization in sleep laboratories with polysomnographic instruments with multiparametric tests. For polysomnographic evaluation the patient must sleep in an unfamiliar environment with several wires attached to the body. For this reason this highly intrusive monitoring setting is sub-optimal. The threshold for a potential patient to go to a practitioner to perform polysomnography can be high, and the capacity to monitor patients in sleep laboratories is limited. The practitioner is often lacking the necessary data to make a proper decision whether polysomnography should be performed. In many cases the partner of the patient is the only source of information.
We aim to accommodate improved diagnosis of OSA by development of new software solutions bridging state-of-the-art consumer electronic devices with appropriate sensors to supplement the classical polysomnography. The form factor of the hardware will be miniaturized and wearable. It is our goal to enable anyone to monitor physiological parameters that are relevant for OSA monitoring at home (or anywhere). This requires that the monetary costs and the efforts needed for monitoring are kept low. As a consequence, the threshold for a first step towards OSA diagnosis should be much lower, i.e., it is much easier to use an app on the smart phone than to order and use specialized medical equipment.
To achieve these goals the project performs interdisciplinary research, with three partners from the medical domain and experts in the areas of Obstructive Sleep Apnea (OSA) and next generation of sensors for medical use; and two partners from computing with expertise in mobile systems, sensor data acquisition and processing, signal processing, data analysis, and event detection. The application requirements are determined by the medical experts that will also perform user studies. An extensible data acquisition system will be implemented with smart phones and sensors, like Shimmer motes and the Bitalino sensor set. This system will be used to collect longitudinal data from sleep monitoring at home and in the sleep laboratory (combined with classical polysomnography to annotate the ground truth).
Supervised learning (data mining) techniques will be systematically studied for their use to automatically analyze longitudinal data for OSA detection. These studies will use data from the PhysioNet databases (early project phase), and later-on data that has been collected in user studies with the data acquisition system. Furthermore, we investigate the usefulness of supervised and unsupervised learning (data mining) techniques to identify interesting data patterns that might lead to new knowledge in OSA research and to support the design and engineering of the on-line analysis tool. The design of the on-line analysis tool is driven by the goal to enable individuals with limited computing skill to customize and personalize the on-line analysis. To achieve this goal, the following three principles will be strictly applied: use of a declarative approach with Complex Event Processing, using few powerful abstractions of physical and logical sensors, and a fine granular modularization implemented in sensor hierarchies. Furthermore, the team will build tools to quantify the quality of off-line and on-line data analysis results.
We aim to enable improved diagnosis of Obstructive Sleep Apnea (OSA) by development of new software solutions bridging state-of-the-art consumer electronic devices with appropriate sensors to supplement the classical polysomnography and to enable:
- monitoring at home with low costs and efforts,
- off-line data analysis of longitudinal monitoring data for research and practitioners,
- identification of new clinical parameters in OSA diagnosis and treatment,
- minimal intrusive,
- seamless upgrade to integrate new sensors,
- on-line data analysis to identify data patterns indicating early warnings for serious health conditions, and
- quantified quality of analysis results.
These new software solutions comprise:
- tools to learn from time-series data from sensors,
- a methodology to reduce the need for computing skills to customize on-line monitoring of OSA,
- tools to quantify the quality of data analysis results, and
- an extensible platform for data acquisition.