Post by Deakin School of Nursing & Midwifery; Centre for Quality & Patient Safety Research

🎥 Video 5 of our 8-part series: Turning 'feedback' into 'foresight': How machine learning is turning thousands of patient complaints into real-time safety interventions. In this video, Dr Muhammad Nouman talks about how advanced analytics and large language models are reshaping the way healthcare organisations understand and act on patient feedback. Key insights from the research: 📌 Classified concerns into core themes - listening, respect, and patient rights - while also identifying the specific healthcare professions involved. 📌 Combined the Healthcare Complaints Analysis Tool (HCAT) with the Patients Come First framework and LLM-powered processing to detect non-verbal cues and generate concise complaint summaries. Using machine learning, the team analysed 13,000+ patient complaints (2018–2024), achieving 97.04% accuracy in automatically detecting clinical communication failures - a major step toward real-time safety triage. Read the abstract 👉 'Identifying Clinical Communication Failures Through Machine Learning Analysis of Routine Patient Feedback' on page 215: https://lnkd.in/g9j4rR6e Stephanie Chappel | Guncag Ozavci PhD, MBA, MSc, BPharm | Cassie McDonald | Sophie Wallace | Mary Dahm | Rebecca Brough CPsychol | Bernice Redley PhD, GAICD | Tracey Bucknall Institute for Health Transformation | Deakin Research | Alfred Health | Alfred Research Alliance #NursingResearch #HealthcareAI #ComplaintClassification #NLP

Video Content