Unlocking the patterns of disease hidden in medical images

23rd August 2016
Visual perception is one area where humans excel, but what if a machine could match or even exceed our ability to discriminate objects and identify patterns in what we see?

Dr Ben Glocker
Reader in Machine Learning for Imaging, Department of Computing

Behind the 2015 victory of an artificial intelligence against Fan Hui, Europe’s reigning champion player of the board game ‘Go’, lies a new generation of deep learning algorithms that are helping computers to perform ever more complex tasks. Visual perception is one area where humans excel, but what if a machine could match or even exceed our ability to discriminate objects and identify patterns in what we see?

Applied to complex medical imaging problems, such as quantifying damage after traumatic brain injury, deep learning methods using artificial neural networks are breaking new ground. AI radiographers – powered by intelligent algorithms – will soon complement human skill, guarding against medical errors and freeing doctors to focus on the most important decisions. But this is just the start – techniques that allow machines to develop a semantic understanding of image data by analysing very large datasets are already showing promise in identifying disease-related patterns and phenomena entirely new to medicine.

I’m curious about…“the new medicine waiting to be discovered in the medical scans we take today”

Bio

Dr Ben Glocker is a Reader in Machine Learning for Imaging at the Department of Computing, Imperial College London. He holds a PhD from TU Munich, and was a post-doc at Microsoft Research Cambridge and a research fellow at the University of Cambridge. Ben is the co-lead of the Biomedical Imaging Analyis (BioMedIA) group at Imperial College. In 2016, Ben was nominated as a World Economic Forum Young Scientist.

Research

Ben’s research focuses on applying machine learning techniques for advanced biomedical image computing and medical computer vision, spanning three disciplines:

  • Medical image analysis;
  • Machine Learning including deep learning;
  • Computer vision.

Ben asks: Can we automatically extract clinically useful information from medical images in order to support clinicians as they work to improve diagnosis, therapy and treatment?