Image Analysis and Research: Building Bridges
Image: A sympathetic neuron, expressing a CFP tagged protein, analysed with Fiji.
A new initiative from the Royal Microscopical Society aims to bridge the growing gap between research and image analysis.
Be it Fiji-ImageJ, CellProfiler or Python scikit-image, in the coming years, the Image Analysis Focused Interest Group of the RMS will improve and extend the use of such methods across biosciences and other fields.
“It's been becoming more and more obvious that there is a growing demand for analysis and for any researcher that's been using a microscope, image analysis is key,” points out Dr Dominic Waithe, acting Chair of the new group and also a UKRI Innovation Fellow at the MRC Weatherall Institute of Molecular Medicine, Oxford, UK.
“Image analysis technology has developed faster than its teaching, and conventionally, the biological sciences have not prioritised mathematics,” he adds. “So we see many researchers, especially in the biosciences, with a poor understanding of analysis; this initiative will get these researchers, and indeed any scientist using microscopy, up to speed.”
Waithe has a wealth of image analysis experience. Having spent many years working with nano-immunologist, Christian Eggeling, on bioimage analysis at the Wolfson Imaging Centre, Oxford, he now develops imaging and analysis methods at Wolfson and the MRC Centre for Computational Biology, University of Oxford.
“The field is moving so fast we can't rely on the university degrees that we did years ago for this, we really need to have more training."Dominic Waithe
However, while taking part in the EU initiative, Network of European Bioimage Analysts (NEUBIAS), he realised how meagre the support was for image analysis in the UK. As he puts it: “I just felt frustrated that there wasn't enough going on here in the UK.”
He soon attended a RMS meeting, where talks with a host of RMS Council members identified the urgent need to address the lack of UK image analysis support. IAFIG-RMS was formed.
Importantly, IAFIG-RMS is different to NEUBIAS. While the EU initiative focuses on developing the interests of bioimage analysis, IAFIG-RMS is set up to include all scientists who are in some way involved in microscopy image analysis.
Dr Dominic Waithe, acting Chair of the Image Analysis Focused Interest Group of the Royal Microscopical Society.
According to Waithe, these researchers could be mathematicians, research engineers, bioinformaticians, biophysicists, microscopists, computer vision scientists or academics from another field. What is important, is the researcher is involved in image analysis.
“The common link really will be image analysis,” he says. “The backgrounds may seem disparate but researchers will all be converging on this topic.”
Growing a community
Since its inception in Spring last year, IAFIG-RMS has carried out a number of events and is set to deliver training courses on bioimage analysis with Python and Fiji-ImageJ. Training courses on CellProfiler and other methods will soon follow.
At the same time, a recent workshop called 'Train the trainer' taught image analysis users the principles of developing their own courses. And at this year's MMC2019, IAFIG-RMS alongside NEUBIAS UK will host the session, UK Applied Image Analysis. As part of this Shoaib Sufi from the UK Software Sustainability Institute is to talk about best practice in software products for use in research.
Learning the principles of delivering training courses at the IAFIG-RMS workshop, 'Train the trainer'.
Clearly, in just over a year, IAFIG-RMS has garnered much interest from a broad range of researchers. And momentum is gathering.
Members of EU bioimage analysis network, NEUBIAS, already view IAFIG-RMS as a model to continue its activities, alongside other organisations, when it ends in 2021. As Waithe points out, IAFIG-RMS would now like to attract more people who are not necessarily in a support role.
“We hope to develop a community of people that focuses on research but also see the benefits of being involved in such a support role,” he says.
Indeed, a recent IAFIG-RMS survey of researchers and facility managers, revealed that nearly 70% of respondents would be interested in a grant that provides support for community training as well as research. As such the group will now approach UK Research Councils for support with a Fellowship that can provide just this.
“We will petition for Fellowships that also have a support role,” says Waithe. “This could allow people from, say, computer vision, machine learning or biophysics to get a benefit from offering support... many are interested in providing support to use the techniques that they develop but this isn't written into their research grants.”
Looking further into the future, Waithe also intends to provide a RMS-endorsed document of best practice for researchers publishing data that will have been produced as a community effort. And an online MSc specifically on image analysis could also ensue.
“IAFIG-RMS has been well-received and I expect that in the future we will have a RMS Committee just for image analysis,” he says. “The field is moving so fast we can't rely on the university degrees that we did years ago for this, we really need to have more training.”