Osteoporosis ("porous bone") is characterized by a loss of bone density that results in a high risk of fractures. The disease is a major cause of pain, disability and death in older people.

Currently, diagnosis involves measurements of bone mass and density using specialized X-ray or CT techniques. In many cases, though, the disease is not spotted until it's reached an advanced stage - i.e. after a fracture has occurred - and the patient requires implants or surgical plates to treat or prevent further fractures.

What the Swiss team has done is to combine density measurements with a large-scale mechanical analysis of the complex inner-bone microstructure (which accounts for the bone's capability to bear loads and therefore represents a better indicator of a bone's true strength).

Using large-scale, massively parallel simulations, the researchers were able to obtain a dynamic "heat map" of strain, which changes with the load applied to the bone. This map shows exactly where, and under what load, a bone is likely to fracture.

"Knowing when and where a bone is likely to fracture, a clinician can also detect osteoporotic damage more precisely and, by adjusting a surgical plate appropriately, can determine its optimal location," explained Costas Bekas of IBM's computational sciences team in Zurich.

The IBM-ETH team was able to conduct the first simulations on a 5x5 mm specimen of real bone by using the massively large-scale computing capabilities of the eight-rack Blue Gene/L supercomputer. In 20 minutes of computing time, the simulations generated 90 Gbytes of output data.

"Ten years from now, the performance of today's supercomputers will become available in desktop systems, making such simulations of bone strength a routine practice in CT," claimed Alessandro Curioni, manager of IBM's computational sciences group in Zurich.

The next task is to advance the simulation techniques beyond the calculation of static bone strength to study the actual formation of fractures in individual patients. That would represent a big step towards fast, reliable and early detection of people with high fracture risk.

The Zurich team detailed its findings in a paper presented at the 5th European Conference on Computational Methods in Applied Sciences and Engineering (ECCOMAS 2008) in Venice, Italy, earlier this week.