Bones are continuously remodeled (resorbed and regenerated) to allow fracture healing and skeleton adaptation to stress. When excessive resorption occurs, bone microstructure is deteriorated, leading to osteoporosis. At early stages, osteoporosis usually has no symptoms; most people are diagnosed when a fracture occurs due to disease severity. To prevent fractures, technologies have been developed to identify high risk population eligible to treatment. Fracture risk has been assessed by analyzing the interaction of different energy stimulus with bone tissues as well as by statistical models that evaluate multiple clinical risk factors. The most applied methods are Dual-energy X-ray Absorptiometry and Fracture Risk Assessment tool. As they present some limitations, other technologies have been proposed for such purpose. A survey of the currently applied and emerging methods is here presented in order to provide a scenario of the technological challenges and trends to diagnose osteoporosis.

In recent years, machine learning (ML) techniques have gained popularity in structural health monitoring (SHM). These have been particularly used for damage detection in a wide range of engineering applications such as wind turbine blades. The outcomes of previous research studies in this area have demonstrated the capabilities of ML for robust damage detection. However, the primary challenge facing ML in SHM is the lack of interpretability of the prediction models hindering the broader implementation of these techniques. For this purpose, this study integrates the novel Shapley Additive exPlanations (SHAP) method into a ML-based damage detection process as a tool for introducing interpretability and, thus, build evidence for reliable decision-making in SHM applications. The SHAP method is based on coalitional game theory and adds global and local interpretability to ML-based models by computing the marginal contribution of each feature. The contribution is used to understand the nature of damage indices (DIs). The applicability of the SHAP method is first demonstrated on a simple lumped mass-spring-damper system with simulated temperature variabilities. Later, the SHAP method has been evaluated on data from an in-operation V27 wind turbine with artificially introduced damage in one of its blades. The results show the relationship between the environmental and operational variabilities (EOVs) and their direct influence on the damage indices. This ultimately helps to understand the difference between false positives caused by EOVs and true positives resulting from damage in the structure.

A significant problem associated with the implementation of Vibration-Based Structural Health Monitoring (VSHM) systems originates from the detrimental effects caused by Environmental and Operational Variations (EOVs). The EOVs cause observations from the same structural condition to behave in different manners. As such, this leads to issues when defining a robust baseline state as well as making the discrimination between undamaged and damaged observations more challenging. In order to address these challenges, multivariate nonlinear regression is implemented to account for the EOVs. Damage Sensitive Features (DSFs) are extracted from acceleration data and then are regressed based on environmental and operational parameters. New features are effectively normalised by finding the difference between the measured and predicted values. This process removes the influence of the EOVs in an explicit manner, allowing for more reliable damage detection. While the benefit of the application of different regression methodologies has already been demonstrated in the past, this work addresses a number of practicalities in the implementation of VSHM on real systems. The analysis investigates the selection of the DSF and is investigated alongside another analysis into how the damage detection behaves under varying amounts of input information. Furthermore, a method is proposed to understand and account for a large number of outliers in the training data.