Predicting and locating fracture in bone using acoustic emission

Abstract

A novel way to predict intra-operative fracture during Total Hip Arthroplasty (THA) using acoustic emission (AE) has been tentatively discovered. AE has also been shown to be able to predict the location of such fracture. This work has potential benefit for the THA surgeon as it gives him a warning of when fracture is imminent and secondly where on the bone it is likely to occur. Eight bovine femora were tested using a Materials Testing Machine. Mock implants were forced into the specially prepared femora until the femora fractured. Both strain guages and AE sensors were mounted on the femora. Strain was used as a control to indicate when the femur fractured. The data from the AE sensors was analyzed post test to determine a parameter that could be used to predict when fracture was imminent and indeed when it had occurred. It was discovered that the peak frequency of the AE waves reduced significantly just before fracture occurred. It is theorized that as the bone material undergoes microcracking, the properties of the materials alter resulting in this change in peak frequency. Two AE source location algorithms were tested on rectangular samples of bone harvested from the mid diaphysis of bovine bone to determine the feasibility of predicting the location of the fracture by locating in real time the microcracks that occur as their prelude. The source location algorithms detected artificial AE sources (pencil lead breaks) to just over 1 mm (on average) of their true location. Then three samples were loaded in three point bending until they fractured. The source location algorithms located the microcracks using AE data collected during the tests. The computed locations showed that there was a close correlation between where the microcracks were detected and where the fractured occurred.