In recent years, artificial intelligence (AI) has increasingly moved beyond diagnostics and automation into a more human domain: understanding emotions. Nowhere is this advancement more urgent than in paediatric care—particularly for children living with multiple long-term conditions (MLTCs) who often suffer silently under the weight of emotional distress.
As a data scientist and academic researcher, I recently led a pioneering study that leverages AI to decode the emotional experiences of chronically ill children using natural language processing (NLP). This research, published in a peer-reviewed medical journal, reveals how AI can shine a light on unspoken suffering and help clinicians, caregivers, and policymakers craft better, more compassionate responses.
The Invisible Side of Paediatric Illness
Children with MLTCs—conditions such as diabetes, epilepsy, cystic fibrosis, and neurological disorders that co-occur—are typically the most vulnerable in the healthcare system. Beyond the physical toll, these young patients often grapple with fear, sadness, frustration, and social isolation.
Unfortunately, emotional distress among children is frequently underreported. Young people may lack the vocabulary or confidence to articulate their struggles, and healthcare systems are seldom equipped to probe deeply into psychological burdens. That’s where AI comes in.
Giving Voice to the Voiceless Through AI
Our research deployed a fine-tuned AI model known as CoEmoBERT, a transformer-based language model capable of interpreting sentiment and emotion in textual data. We collected thousands of anonymised posts from social media platforms and online forums where children, parents, and caregivers shared narratives about paediatric chronic illness.
What we found was both heart-wrenching and illuminating: across six core emotions—sadness, fear, disgust, amazement, satisfaction, and neutrality—sadness and fear dominated. These emotions were more common in cases involving multiple overlapping conditions and long treatment durations.
The AI model could not only detect emotional tone but also quantify and cluster sentiments by illness type, gender, and perceived treatment success. This level of granularity would have been nearly impossible through manual analysis alone.
Why This Matters
Understanding emotion in healthcare isn’t just a ‘nice to have’—it’s a clinical imperative. Poor emotional wellbeing among children with chronic illnesses is linked to lower treatment adherence, diminished quality of life, and higher healthcare costs. By applying emotion-aware AI, healthcare systems can flag mental health risks earlier and tailor care more empathetically.
For instance, an AI-powered dashboard could alert clinicians when online posts by a patient indicate rising anxiety or despair, enabling timely psychological support. At scale, such technology could complement electronic health records and even shape national child health strategies.
A British Innovation with Global Impact
This project represents not just academic excellence but applied innovation in line with the UK’s digital health vision. The model was developed during my MSc in Data Science and Business Analytics at the University of Plymouth and has now been nominated for further dissemination through multiple journals and conferences.
As someone in the tech industry, I believe this research illustrates the kind of forward-thinking innovation that blends artificial intelligence with real-world impact. It reflects the UK’s ambition to lead globally in ethical AI and human-centred tech.
Toward a More Emotionally Intelligent Healthcare System
The future of AI in healthcare isn’t merely about predicting diseases—it’s about understanding patients. Emotion-aware AI offers a bridge between cold data and human experience. It allows us to listen to children even when they don’t speak.
This is the essence of true innovation: using technology to deepen empathy, not just efficiency.
As the UK welcomes global talent to solve its most pressing challenges, it’s time we recognise the power of AI to not just save lives—but to understand them.
About the Author:
Temidayo Oluwalade is a data scientist, educator, and published researcher in AI for healthcare. He holds an MSc in Data Science & Business Analytics from the University of Plymouth.
