All Posts By

M3i

M3i und Ernst & Young present “Pulse of the Industry” event, 10/22/18, Munich

By | News

Together with Ernst & Young, one of the leading audit firms, M3i presents this year’s “Pulse of the Industry” event. Leading experts from the field of medical technology will give presentations on topics relevant to medtech companies today.

When & where:
October 22nd 2018, 4pm
Hörsaal der Poliklinik, Pettenkoferstr. 8a, 80336 Munich, Germany
Registration: lh@m3i-muenchen.de

M3i research project VibroSuite to be presented at CAOS 2018 in Beijing

By | News

We are excited to announce that VibroSuite, one of our recent research projects, has been accepted as a contribution at this year’s CAOS International Conference in Beijing. Nima Befrui of LMU München is speaking on Thursday, June 7th at 08:33 to present our work on “Detection and Grading of Knee Joint Cartilage Defect Using Multi-Class Classification in Vibroarthrography”.

An article about the project was also published earlier this month in a peer-reviewed journal (see here for the full article).

M3i and Munich Innovation Labs co-publish study on Vibroarthography with clinical partners

By | News

M3i and its partners, Munich Innovation Labs and the Munich University Hospital are excited to announce the publication of their first co-authored article, “Vibroarthrography for early detection of knee osteoarthritis using normalized frequency features” at Springer Nature (see here for a free, full length read-only version). The article discusses the findings from research project “ArthroScan”, conducted by researchers from M3i and the Munich University Hospital. In this EU-funded project, researchers sought to study the possibilities of vibroarthography, a non-invasive, non-radiative and very cost-effective technology using sound waves for early-stage detection of damage in knee joints. Munich Innovation Labs contributed its expertise in machine learning to the project, to develop a technology based on machine learning and a linear support vector machine, which proved to reach a classification specificity of approximately 0.8 at a sensitivity of 0.75. This performance is comparable to existing diagnostic tools, thus qualifying machine-learning- supported vibroarthography as an additional diagnostic tool.