Detection of Bowel Sounds
Abstract
This thesis describes the development and implementation of a bowel sound detectionalgorithm, which ultimately might be used as part of a meal detection system in an artificialpancreas. An artificial pancreas is an advanced blood glucose regulation system for patients with diabetes. In order to correctly provide insulin doses to the user, artificial pancreases need to detect when the patient is eating. One possible approach for meal detection is monitoring bowel sound occurrence frequency. This further necessitates a bowel sound detection system, such as the one described in this thesis.
The detection algorithm will consist of a combination of signal processing techniquesfor feature extraction, and pattern recognition in order to classify sound segments as eitherbowel sound segments or non-bowel sound segments. A linear support vector machine waschosen for classification in the final solution. The Python implementation of the detectionsystem yielded promising results on a provided data set of sound measurements from thestomach. However, certain improvements and refinements should be done to the systembefore potential integration into an artificial pancreas. The thesis also briefly discussesthese further development needs.
Chapter 1 explains the motivation behind the assignment, as well as the problem descriptionand approach. Chapter 2 provides the most important theoretical background for thework described in this thesis. Chapters 3 and 4 describe the method and work towards thefinal product. These chapters also contain intermediate results which constitute the basisfor decisions made throughout the process. Results obtained from the final solution can befound in chapter 5, while chapter 6 summarizes and concludes the report.