Medical imaging refers to several different technologies used to view the human body and its organs or tissues to diagnose, monitor, or treat medical conditions. Commonly used methods include medical X-ray imaging, computed tomography, magnetic resonance imaging (MRI), and ultrasound. ML involves creating algorithms that automatically improve with experience, and relies on pattern recognition to do so. Hence, ML is naturally suited for identifying patterns, which can be well utilized for medical image analysis. In more and more cases, ML systems are identifying patterns that are beyond human perception, so the use of ML in medical image analysis and visualization for computer-aided diagnosis (CAD) is continually increasing.

This lecture will outline the fundamental ML processes involved in medical image analysis. Achieving prediction and classification for CAD applications will also be discussed. Some preliminary ideas of 3D reconstruction and viewing as applied in medical image analysis will also be presented.

Patterns of diseases are changing; they relate less and less to sudden infections or crippling accidents, and on a growing scale they develop as slow and debilitating afflictions caused by repetitive harmful behaviours (e.g., poor nutrition). These behaviours relate to different domains—the physical, psychological, social and environmental—and contribute significantly to the individual’s overall health and Quality of Life (QoL) in the long term.

The webinar will present research on overall health and life quality and focuses on a robust, reproducible and evidence-based modelling approach based on reliable, privacy-preserving research infrastructure, leveraging a “Living Lab” approach to capture ‘small personal data’ acquired throughout everyday life via an individual’s smartphone and wearables. The modelling approach integrates a mix of theoretical components coming from computer science (pattern recognition, machine learning, deep learning, computational sensing, and artificial intelligence) and theories and models from social and behavioural sciences and preventive medicine.