Ultrasound Biomarkers and Machine Learning

Ultrasound imaging today includes a large panel of quantitative techniques measuring complementary biological aspects of tissues. This leads to an unprecedented richness of information in the data that can be obtained by a single ultrasound examination. In this talk I will explain how we aim to make sense of this data, and extract the most relevant information for a given application, using concurrent ultrasound imaging and machine learning analysis. This will be illustrated with studies investigating potential new biomarkers in breast cancer diagnosis, and exploring the neuronal activity patterns associated with specific behavior in rodents and non human primates.