This page contains a Flash digital edition of a book.
MODELLING: BIOLOGY


The computer as healer


The overriding goal of human biological modelling is to help us get healthy and keep us that way, and this can take shape in an immense variety of forms. Paul Schreier examines some interesting projects where computer models play a key role


Improving workplace ergonomics, studying why epileptic seizures occur, trying to identify and prevent future HIV epidemics, designing medical implants that duplicate or augment the natural response of body tissues – these are just a few of the many areas where computer modelling is helping to make our lives better.


The virtual soldier comes home The entire human muscular system is being modelled in an unusual way at SantosHuman Inc. Leveraging research from the Virtual Soldier Research programme at the University of Iowa, they have created Santos. He is a computerised avatar designed with a complete biomechanical muscular system to provide feedback on fatigue, speed, strength and torque. Santos performs actions in the virtual world to address needs across a wide variety of industries. Ergonomists can search for better designs; the military can provide greater safety and performance for soldiers; those creating amusement-park attractions can determine the forces and effects on riders; while manufacturers can provide improved human-interface requirements and reduce costs while increasing worker safety. For instance, Santos has joined the workforce


at Ford Motor Co. and will help manufacturing engineers test ergonomics and safety on a virtual assembly line and thereby help create the safest and most ergonomic way to build a vehicle.


16


For example he can perform a task and tell whether, over months and years, it will cause back strain – and the car manufacturer can make adjustments until they find the optimal way to get the job done. With the software users can see body strength in real time, see fatigue plus observe and predict motion. You can get a visual idea of what Santos can do by going to his Facebook page (search on ‘Santos VersionOne’) and viewing the various videos posted there. The software is based on predictive dynamics, what the developer describes as a novel approach for simulating human motion. The basic idea is to model a redundant dynamic system as an optimisation problem and solve for its motion where only limited information about the system is available. Both


the motion and forces that cause the motion are unknown and treated as design variables in the optimisation process. Equations of motion are treated as equality constraints instead of their direct numerical integration. The simulation combines all the robotic formulas with physics equations to describe 216 degrees of freedom covering every joint in the body, explains Jay Johnson, CEO of SantosHuman. He adds that it takes from 20 minutes to five hours to build a simulation, and a simulation calculation run can then take from 10 minutes to half an hour. But, once the predictive dynamics algorithm is defined, the simulation can run continuously in real time.


How large must the model be? To better understand why epileptic seizures occur and propagate, scientists at the Argonne National Laboratory have created a model of small areas in the brain using neural networks. Traditionally in such models, each neuron has been treated as a simple entity existing in only one of two states (firing or inactive), but this more sophisticated model treats each neuron as a pathway unto itself; it traces the route of an electrical signal from the fibrous dendrites into the cell body and out through the axon to other neurons. In other words, it treats a neuron as a data chain, where each link represents a different physical site on the cell. The software for this task has been evolving,


When mounting tyres in a car factory, the Santos avatar can describe all the forces on his joints to predict stress and fatigue. (Source: Santos VersionOne Facebook page)


explains experimental systems engineer Mark Hereld. A first form of neural modelling uses an open source code called GENESIS (GEneral NEural SImulation System). With it scientists can build single or multiple cells on a desktop PC or workstation. One problem, though, is that the software is not scalable. Given a simulation with 100,000 elements, the program starts to trip over itself when passing data among the elements and performing housekeeping.


SCIENTIFIC COMPUTING WORLD AUGUST/SEPTEMBER 2010 www.scientific-computing.com


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48
Produced with Yudu - www.yudu.com