Modeling reality

Mathematical models are convenient ways for scientists to represent, understand and predict what happens in the world outside of the lab. But any model is a simplification of what it represents, and we need to ask: how closely do models fit the real world?

For the Blast research group at Imperial College, London, creating a realistic model of the human body, and in particular the legs, is a vital part of research. Despite being a difficult, complex process, the group has created some of the most sophisticated models used in research today.

Understanding the composition of the millions of different elements of a single bone, and how they respond to the forces they are subjected to is key to understanding the dynamics of an injury.

However, a model is only as good as its inputs and programming. An older man with osteoporosis will have a dramatically different bone make-up than a young athlete. A previously damaged and healed bone may have unpredictable weaknesses. While the Blast Lab has accounted for some of the real world variability in their models, their estimations are not perfect.

The ultimate aim for the group – or any model creators dealing with complex and chaotic systems, from medicine to climate science – is to devise something that represents the real phenomenon accurately and generates the information needed to make useful, practical decisions. The level of detail they’re seeking may seem pedantic, but with every small modeling advancement, their goal of understanding, and therefore preventing, blast injuries becomes to be a little easier to reach.