This virtual course will show how public bioimaging data resources, centred around the BioImage Archive, enable and enhance machine learning based image analysis. The content will explore a variety of data types including electron and light microscopy and miscellaneous or multi-modal imaging data at the cell and tissue scale. Participants will cover contemporary biological image analysis with an emphasis on machine learning methods, as well as how to access and use images from databases. Further instruction will be offered using applications such as ZeroCostDL4Mic, ilastik, the BioImage Model Zoo, and CellProfiler.

Virtual course

This course will be a virtual event delivered via a mixture of live-streamed sessions, pre-recorded lectures, and tutorials with live support. We will be using Zoom to run the live sessions (all fully password protected with automated English closed captioning and transcription) with support and both scientific and social networking opportunities provided by Slack and other methods, taking different time zones into account.

In order to make the most out of the course, you should make sure to have a stable internet connection throughout the week and are available between 08:00 – 18:00 BST each day. In the week before the course there will be a brief induction session. Computational practicals will run on EMBL-EBI's virtual training infrastructure, meaning participants will not require access to a powerful computer or install complex software on their own machines.

Selected participants may be sent materials prior to the course. These might include pre-recorded talks and required reading or online training that will be essential to fully engage with the course.

Who is this course for?

This course is aimed at scientists working with biomage data across the life sciences. It is suitable for those involved in creating bioimages or taking their first steps in analysis. The content is also suitable for those wanting to learn more about the BioImage Archive and gain experience with machine learning approaches for image analysis. The programme will be of particular interest to bioimage analysts with questions relating to the use of ‘big data’ and using the wealth of publically available data curated in the BioImage Archive.

The course is accessible to members of the bioimaging community and does not require prior experience with machine learning methods or use of the BioImage Archive. Applicants are encouraged to explore the resources below before starting their application. Applicants should be comfortable with basic programming tasks and have experience working with Python.

Recommended preparatory reading:

• BioImage Archive: A call for public archives for biological image data
• ZeroCostDL4Mic: an open platform to simplify access and use of Deep-Learning in Microscopy
• EMPIAR: a public archive for raw electron microscopy image data
• Image Data Resource: a bioimage data integration and publication platform
• BioImage Model Zoo

What will I learn?

Learning outcomes

After the course you should be able to:

• Interact programmatically with the BioImage Archive and other data resources
• Apply pre-built machine learning models to image data
• Train and retrain machine learning models on image data
• Utilise machine learning approaches for object detection, image segmentation, and de-noising
• Generate quantitative conclusions from images

Course content

During this course you will learn about: 

Data repositories

• BioImage Archive
• Image Data Resource
• EMPIAR
• BioStudies

Analysis tools

• ilastik
• Colaboratory
• BioImage Model Zoo
• ZeroCostDL4Mic
• ZeroCostDLMicEverywhere
• CellProfiler
• Cellpose

This course is currently under development, please check back again soon for further details.

Trainers

Please read our support page before starting your application. In order to be considered for a place on this course, you must do the following:

• Complete the online application form.
• Ensure you add relevant information to the ‘submission details’ section where you are asked to provide information on your:
  • pre-requisite skills and knowledge
  • current work and course expectations
  • data availability
• Upload one letter of support from your supervisor or a senior colleague detailing reasons why you should be selected for the course.

Please submit all documents during the application process by 23:59 on 4 February 2024. Items marked * in the application are mandatory. Incomplete registrations will not be processed. 

All applicants will be informed of the status of their application (successful, waiting list, unsuccessful) by 20 February 2024. If you have any questions regarding the application process please contact Meredith Willmott.

Course materials

The course materials from the 2023 edition of the course are live and available for you to use. They provide a mixture of pre-recorded lectures, presentations, and practicals from the course, and will give you a snapshot of what to expect in the 2024 edition.

Flash talks

All participants will be asked to give a short presentation about their research work as part of the course. These provide an opportunity to share their research with the other participants and provide a forum for discussion. Successful applicants will be required to submit their talks upon registration.

Financial assistance

This course has been included as part of the Professional Development Programme of Bioimaging North America (BINA). Successful applicants to the course from North-American countries (Canada, US, Mexico) who are members of BINA can apply directly to BINA to get financial support to cover this course’s registration fee. Please contact BINA before registration as reimbursement of the course fee will not be possible. While virtual formats allow us to reach a wider geographic audience please note that this training event will run on BST time zone.

AI4Life

Some of this course’s organisers and trainers are partners of the AI4Life Project. AI4Life has received funding from the European Union’s Horizon Europe research and innovation programme under grant agreement number 101057970.