Better Imaging of Living Animals and the 3Rs

Recently, I was explaining my research to another academic from a different field. I was describing how, when imaging living animals under the microscope, we strive to keep laser power levels to the minimum needed. The question I got in return was: “Is that for the animal’s sake or for the imaging?”. And of course the answer is both - in general, I find that good science and better microscopy align very well with performing more humane animal research.

This article is about how researchers are working to improve animal research by minimising numbers used, sharing animals and data and refining experimental procedures.  This is not an article on whether or not animal experimentation should or should not be done; that is a matter of personal ethics. However, should you be interested in the rules and regulations surrounding animal research in the UK, I would suggest you start with Understanding Animal Research [1] and the National Centre for the Replacement, Refinement and Reduction of Animals in Research (the NC3RS) [2].

Principles of the 3Rs

The 3Rs of Replacement, Reduction and Refinement having been guiding principles for the ethical use of animal research for almost 60 years now [3]. There are many places that state definitions but, for ease, these are the definitions as used by NC3RS:

Replacement - Accelerating the development and use of models and tools, based on the latest science and technologies, to address important scientific questions without the use of animals.

Reduction - Appropriately designed and analysed animal experiments that are robust and reproducible, and truly add to the knowledge base.

Refinement - Advancing research animal welfare by exploiting the latest in vivo technologies and by improving understanding of the impact of welfare on scientific outcomes.

All three of these principles are important to better science and collecting better data.


Reduction is perhaps the simplest of these three principles and the one that most clearly relates to excellence in science. Reduction refers to approaches that minimise the number of animals used in research. Importantly, reduction refers to using the minimum number of animals needed to answer the scientific question under study.

Reduction can be accomplished by all researchers by incorporating best practices in experimental design, analysis and statistics [4]. This also ensures robust, reproducible findings - a central tenet of best scientific practice.

Many projects aimed at reducing animal numbers take the form of improving longitudinal studies - studies that take measurements at several time points. With many approaches such studies require the killing of N animals per time point. By incorporating state-of-the-art imaging technologies we are able to image each animal at all time points without killing animals; thus, the number of animals drops from tN, where t is the number of time points, to just N.

Using the same animals at each time point also helps to reduce the confounding effect of between-animal variation. Animals, like humans, often show wide variation between individuals and using a completely distinct sample of animals at each time point can obscure small temporal effects.

The use of advanced imaging can also reduce animal numbers by allowing multiple organs or tissues to be imaged within a single animal. For example, one might be interested in the effect of a new cancer drug on cancerous and healthy tissues within an animal. With clever, non-invasive imaging approaches this can be achieved with minimal concern that the acquisition of one dataset affects the other.

Further, imaging experiments can produce massive swathes of data. With appropriate infrastructures, scientists can share these datasets and allow other scientists to re-use and re-analyse the data and also combine their data with other datasets to help answer scientific questions without any further animal research.


But, as scientists, we must be careful that reduction of numbers is balanced with refinement of experiments.

Refinement is a bit more difficult to measure and there may be totally unique refinements needed for a particular animal experiment. Refinement refers to the modification of experimental design such that experiments still answer the scientific question in mind whilst minimising both short-term and long-term pain, suffering and distress in the animals use. Refinement can occur out of the lab, e.g. at the husbandry stage, or during experimental procedures.

When we use high power light to image animals, particularly over long time periods, we face issues relating photodamage and photobleaching (see my earlier post). In order to get reliable imaging data we must minimise various factors relating to light exposure and powers. But these also have a distinct effect on the welfare of the animal being imaged. Continuous, high power exposure to laser light, particularly blue/UV light can be very damaging to the animals we’re imaging. This leads to unreliable data, which we don’t want.

Obviously, this is a clear place where the aims of good science and the principles of the 3Rs align perfectly. Intense light exposure can cause short-term pain and distress, particularly in a restrained animal, and negative impacts on our data.

Similarly, we must be very careful with our use of anaesthetics and analgesics, which we often use to minimise distress, suffering and pain during experimentation. These drugs may minimise suffering during the experiment but some anaesthetics have been shown to be distressing when introduced to, for example, aquarium water [5]. Using these drugs may also introduce a confounding effect on the data being taken, for example - how do anaesthetics affect neuronal signally?. Here we find an area where joint research needs to be carried out to improve experimental Refinement but also ensure we understand the confounding effects of these drugs.


And last, but definitely not least, we come to the Replacement of animals by either animals that we consider not to experience suffering, e.g. insects, or by completely replacing animal models with humans, cell cultures or computational models.

Often, the experiments that scientists must do to answer questions about such things as disease, immune response and drug development cannot be carried out in humans for a variety of ethical, practical and scientific reasons. Similarly, when probing questions at the very frontiers of science, computational models may not exist. Further, cells and tissues often behave differently in synthetic scenarios than in living organisms. What this means is that sometimes, to answer certain questions, research must be done with living animals. It should be noted that research in animals also has its own challenges, which sometimes make it an unsuitable route to take.

Actually, research in microscopy and bioimaging, particularly for biomedical research is often about using cell cultures and tissue models and much research uses human-derived cells and tissues. However, a fair part of biomedical research, particularly, for example, neuroscience, required the imaging of live animals. Here imaging isn’t, in the short-term, aligned with the principles of the 3Rs.

However, many imaging techniques are able to gather masses of data. By making this data freely available, microscopy experiments done now can be used to build the mathematical and computational models of the future. Thus, current imaging experiments in living animals can contribute to the Replacement of animals in future years.

Concluding Remarks

As mentioned at the beginning of this article, this is not an article on the ethics of animal research but I hope that this broad overview of how the principles of Replacement, Reduction and Refinement of animals in research aligns with the central drive of many researchers: better science. One day, hopefully in the not-to-distant future, we will be able to move away from animal experiments altogether but, until that time, we strive to make these experiments as non-invasive and as stress-free for our animals as we can.


  1. Understanding Animal Research (
  2. NC3RS (
  3. W. M. S. Russell, R. L. Burch and C. W. Hume (1959) The principles of humane experimental technique. (Sadly, I have been unable to find an open access copy of this text.)
  4. N. Percie du Sert, I. Bamsey, S. T. Bate, M. Berdoy, R. A. Clark, I. Cuthill, D. Fry, N. A. Karp, M. Macleod, L. Moon, S. Clare Stanford and B. Lings (2017) The Experimental Design Assistant. PLoS Biol 15(9): e2003779.
  5. G. D. Readman, S. F. Owen, J. C. Murrell and T. G. Knowles (2013) Do Fish Perceive Anaesthetics as Aversive? PLoS ONE 8(9): e73773.
Chas Nelson
LKAS Research Fellow in Data Science

An interdisciplinary scientist with a background in quantitative microscopy and bioimage analysis.