Medsensio AS - Machine learning for automatic detection of lung sounds
Johan Fredrik Ravn and Anna Dranovska both have Master’s degrees from UiT, Ravn in Computer Science and Dranovska in Business Creation and Entrepreneurship.
What is the customer problem and how can it be solved?
Respiratory sounds are generally examined using a stethoscope, but it can be difficult to detect and classify sounds accurately. Lung auscultation is currently taught to medical and nursing students one-on-one, with each student receiving instruction from an experienced doctor/nurse. This is a resource-intensive practice and the amount of time available for training is limited. The risk of misinterpretation is significant, as is the risk of not detecting lung sounds that may be caused by disease. An increase in pulmonary problems worldwide makes it critical to be able to interpret lung sounds correctly. There is a need for smart solutions that can automatically detect, register and classify respiratory sounds to help doctors, nurses and others. Our solution is based on a computer-based learning platform for automatic detection and classification of lung sounds. Our first training tool will help students in health education programmes as well as nurses and doctors to learn how to classify lung sounds.
Why did you choose entrepreneurship and Medsensio?
It is very exciting to have the opportunity to bring our idea to the market. It is motivating for us to be developing a product that can help people with lung problems. We wanted to focus on entrepreneurship because it gives us a chance to create something new.
What will support from the Research Council mean for you?
Receiving support from the Research Council makes a significant difference. It allows us to further develop the project into a viable product. Now we can make contact with partners and conduct tests with potential users.
What knowledge does the project build on and how is UiT involved?
The project is based on technology developed as part of Johan Fredrik Ravn’s Master’s thesis in Computer Science at UiT. His Master’s project involved data compiled as part of the seventh survey of the Tromsø Study (Tromsø 7) and expert assistance in interpreting the data by pulmonary specialists from the Medical Humanities research group under the Department of Community Medicine at UiT. The inspiration for our first product was to meet challenges experienced by nursing instructors and students at the Faculty of Health Sciences at UiT.