EMBL Course: Deep learning for image analysis

The event information is correct at time of publishing, but subject to changes and cancellations. Please visit the event website for the most up to date information. 

The advent of deep learning has brought a revolution in the field of computer vision, including most tasks and research questions concerned with microscopy image analysis. Neural networks have been successfully used for image restoration, classification and segmentation, for the detection of objects and characterisation of their morphology, for high-throughput imaging and large-scale processing in 3D. Despite these advances, training and deployment of such neural networks remains difficult for practitioners of image analysis. The aim of our course is to close this gap and teach the participants – in the most hands-on way possible – to apply deep learning-based methods to their own data and image analysis problems.

Audience

This course is aimed at both core facility staff and research scientists.

Prerequisites for this workshop are programming skills in Python and ideally Tensorflow, Keras or Pytorch as well as basic knowledge of machine learning theory.

Learning outcomes

After this course you should be able to:

Understand the fundamentals of machine learning methods suitable for image analysis

Advise users/colleagues in strategies to obtain ground truth

Train and use a CNN for a bioimage analysis task studied in the course

Perform simple quality control on the results