Aircraft are large and complex machines, yet even the tiniest crack in the remotest or hardest-to-reach corner can have major consequences in terms of safety and flight worthiness. This makes periodic inspection and maintenance not only vital, but also laborious and time consuming as certain components are concealed beneath layers of other components. The 'Neural net based defect detection system using LRU technology for aircraft structure monitoring' (SELF-SCAN) project developed a technique using guided wave technology to make inspections and maintenance more efficient while enhancing safety. Unlike other approaches to monitoring complex structures, guided wave technology provides large area coverage from a limited number of sensors. However, aircraft structures as well as the environment in which they interact are complex. Detecting defects from the plethora of geometric data collected using guided ultrasonic waves is therefore an incredibly challenging task. Financed by the EU's Seventh Framework Programme (FP7), SELF-SCAN came up with the novel idea of using neural network systems using permanently installed sensors to enable in situ detection. With a consortium drawn from six EU Member States, the project team created an advanced integrated system for structural health monitoring and impending failure detection. The prototype system demonstrated its ability to differentiate between sound and defective components, as well as to detect minute but critical cracks in regions considered inaccessible to other sensors. Once developed further into a commercialised system, ultrasound detection will help bolster safety, lower the risk of catastrophic failure, reduce costs and increase the service life of aircraft components.