Increased incidence of obesity and longer life expectancies will place increased demands on load bearing joints. In the present work, a method of pre-clinical evaluation to assess the condition of the joint and potentially inform on cases of joint deterioration, is described. Acoustic emission (AE) is a non-destructive test methodology that has been used extensively in engineering for condition monitoring of machinery and structures. It is a passive technique that uses piezoelectric sensors to detect energy released from internal structural defects as they deform and grow. The technique has been used with some success in the past to identify characteristic signals generated from the knee joint during activities such as standing and sitting, in candidate arthroplasty patients (1,2). In this study, 40 asymptomatic subjects had AE data generated from their knee joints analysed. Subject characteristics such as age, gender, and lifestyle were disclosed and evaluated against the AE data. Each subject was invited to take a seated position and a piezoelectric AE sensor (Pancom P15, 150kHz resonance, 19mm diameter) was attached to the subject's knee using a wax couplant and tape as close to the articulating surface and on a bony prominence to avoid signal attenuation in the soft tissue. Subjects were invited to sit and stand 3 times. AE data were collected and processed using an AMSY5 AE processor (Vallen, Germany). Tests were repeated on a separate occasion and selected subjects were invited to participate on a third occasion. The AE data of particular interest were the peak amplitudes and the frequency power spectrum of the waveform. Post-test inspection of subject characteristics allowed them to be separated into three broad categories: no previous history (group A), some instances of pain in the knee (group B), and those who have had previous minor surgery on the knee (group C). The corresponding AE results were grouped separately. It was found that groups A and B demonstrated similar signal amplitude characteristics while group C produced much higher, significantly different (p<0.05) amplitudes and amplitude distributions. Typical results are shown in figure 1. At present, broad trends could be identified and relationships emerged between the data and subject history (prior surgery, typical daily activity). Further work will continue with asymptomatic subjects and the work will be extended to pre-operative patients to identify whether certain trends are amplified in this population.
