Full body simulations to predict patients’ response to treatment

Doctors could one day predict how treatments will affect a patient’s entire body using personalised computer models under development at a new research centre.

Academics at Sheffield University’s Institute for Biomedical Imaging and Modelling (INSIGNEO), which launched last week, plan to develop more detailed simulations of different organ systems that could eventually be combined into a single model.

These simulations, built using scans of a patient’s anatomy and data on how it operates — for example, blood flow dynamics — could enable doctors to see how an individual might react to a specific treatment or technology.

The researchers first plan to add information on lifestyle and environmental factors to their existing models to make them more accurate for each patient. They also hope to start demonstrating how treatment of one part of the body can affect other systems.

‘A scan gives you a very static view of the patient today,’ institute director Prof Alejandro Frangi of the university’s mechanical engineering department told The Engineer.

‘What you would like to do is know what sort of exercise the patient does, what are the changes in heart rhythm and blood pressure in a normal week for that individual.

He added: ‘It’s not the same to treat a patient who is a 60-year-old woman, who is retired and has a quiet life, and a 35-year-old man who still has to work and is active in society.

‘And it’s not the same if this man is in a small town somewhere or is a financial officer in a big company in London. The stress through which this person goes and their habits are going to be completely different.’

The researchers also hope to add different levels of detail to the simulations, observing what happens at the molecular, cell, tissue, organ and system level, and at different timeframes.

To do this they need to simplify the simulation at each level so that the calculations can be performed with available computing power but without significantly reducing the accuracy of the model.

Frangi compared this to creating a version of Google Earth for the body. ‘Google Earth allows you to zoom in and out,’ he said. ‘You have different layers of information that are somehow fused into a coherent system.’

However, he stressed that a full-body computer simulation that can show the effects of a treatment on every system as they interact with each was a long-term vision that would take many years to achieve.