Cardiovascular disease (CVD) remain a leading cause of deaths worldwide, with many diagnoses occurring only after the onset of symptoms. This reactive, symptom-driven approach may hinder timely interventions, emphasizing secondary prevention -- managing conditions after they arise -- over proactive primary prevention. Primary prevention, which involves identifying at-risk individuals before symptoms manifest and empowering them to act on modifiable risk factors, such as lifestyle and diet, has the potential to significantly reduce cardiovascular events and alleviate healthcare burdens [3]. Achieving this, however, requires innovative frameworks that integrate personalized data, predictive models, and actionable insights to address the complexities of cardiovascular health management. One promising innovation for advancing primary prevention is the concept of health digital twin. Health digital twins provide virtual representations of patient-specific conditions, enabling personalized risk assessments, outcome predictions, and scenario exploration. By integrating predictive modeling with patient-specific data, this concept offer the potential to shift CVD prevention from a reactive approach to a proactive, data-driven strategy. However, implementing this type of digital twin at scale for CVD prevention poses several challenges [37, 11, 36], among which: - Data integration and harmonization: Health data is often distributed across various systems and formats, which requires a data unification procedure when attempting to integrate it into a cohesive predictive framework.

Deep Learning methods are notorious for relying on extensive labeled datasets to train and assess their performance. This can cause difficulties in practical situations where models should be trained for new applications for which very little data is available. While few-shot learning algorithms can address the first problem, they still lack sufficient explanations for the results. This research presents a workflow that tackles both challenges by proposing an explainable few-shot learning workflow for detecting invasive and exotic tree species in the Atlantic Forest of Brazil using Unmanned Aerial Vehicle (UAV) images. By integrating a Siamese network with explainable AI (XAI), the workflow enables the classification of tree species with minimal labeled data while providing visual, case-based explanations for the predictions. Results demonstrate the effectiveness of the proposed workflow in identifying new tree species, even in data-scarce conditions. With a lightweight backbone, e.g., MobileNet, it achieves a F1-score of 0.86 in 3-shot learning, outperforming a shallow CNN. A set of explanation metrics, i.e., correctness, continuity, and contrastivity, accompanied by visual cases, provide further insights about the prediction results. This approach opens new avenues for using AI and UAVs in forest management and biodiversity conservation, particularly concerning rare or under-studied species.