377 - From Paper to Precision: Faster and More Accurate Patient-Centred Dementia Diagnosis using AI

Dementia is one of the most pressing global health challenges, with devastating impacts on individuals, families, and healthcare systems. It is a leading cause of disability among older adults, contributing to emotional and financial burdens. In the UK, someone develops dementia every three minutes, affecting over 944,000 people (NHS, 2024). Despite its prevalence, diagnosing dementia—especially in the early stages—remains difficult (WHO, 2024). Alarmingly, 36% of cases go undiagnosed, with rates as high as 50% in some regions (Dementia Commission: 2023 report).

Early diagnosis is critical, enabling access to timely support and adjustment to the condition, while maximising the benefits of available medications for Alzheimer’s disease (Dementia Commission Report, 2023). However, barriers persist. According to the Alzheimer’s Society, 52% of patients delay visiting their GP until symptoms worsen. Additionally, the National Audit of Memory Assessment Services in England and Wales (2021) reported that average waiting times for diagnosis increased from 13 weeks in 2019 to 17.7 weeks in 2021, with some regions experiencing delays of up to 104 weeks.

Our vision is to revolutionize dementia diagnosis through machine-learning-powered tools that are accurate, non-invasive, cost-effective, and user-friendly. By enhancing traditional methods like the clock-drawing test with a tablet-based app, we can capture richer datasets for reliable early detection. During the OSC funding phase, we will develop a tailored interface to support older adults, individuals with early cognitive decline, and healthcare professionals in gathering the dataset needed, to create this transformative diagnostic solution.