Key points are not available for this paper at this time.
Purpose of review The management of immunoglobulin E (IgE)-mediated allergic diseases and allergen immunotherapy (AIT) is complicated by high interindividual variability and the unavailability of reliable predictive biomarkers. Given the complexity of the immune response underlying these processes, integrating artificial intelligence (AI) could revolutionize the clinical management of these patients. Recent findings Recent literature highlights how machine learning (ML) can efficiently stratify patients by analyzing omics data and molecular sensitization profiles. Models such as long-term memory recurrent neural network and stochastic latent actor-critic enable the prediction of therapeutic adherence, while deep learning (DL) enables the integration of unstructured data from wearable sensors and environmental parameters for real-time monitoring. At the same time, federated learning (FL) is emerging as a crucial solution for multicenter research, ensuring data privacy in compliance with international standards and regulations. Conversely, although generative AI supports the synthesis of clinical records, recent studies highlight its limitations in terms of reliability for direct patient use. Summary Integrating AI into AIT management could lead to even more efficient precision medicine, improving diagnostic accuracy and predicting therapeutic outcomes. However, while AI serves as a fundamental decision-making aid, specialist supervision remains essential to ensure patient safety and to interpret complex data.
Gangemi et al. (Tue,) studied this question.