Topic of the invited talk: The rapid digitisation of healthcare, propelled by advancements in Artificial Intelligence (AI) and Machine Learning (ML), has revolutionised the management and analysis of Electronic Health Records (EHRs). Among the diverse techniques emerging from this technological frontier, Natural Language Processing (NLP) stands out for its ability to unveil hidden patterns, insights, and the nuanced dynamics of medical professional-patient interactions. Specifically, topic modelling and sentiment analysis provide powerful tools for deciphering the latent themes and emotional undertones in clinical documentation, offering unprecedented opportunities to understand and improve patient care. This is even more important in FND care, as FND patients often feel unheard and stigmatised by clinical professionals. We conducted an NLP-based and sentiment analysis of written records from neurologists, psychologists and other medical professionals supporting FNF patients. The results indicated remarkable differences in the type of vocabulary used and in the ‘emotional tones’ conveyed by different types of professionals when writing to and about their FND patients. Our findings have direct clinical implications and offer insights into how to develop better care pathways for people with FND and how to improve clinical training for professionals working with FND patients.