Carina King, senior researchers at KI presents: I for respiratory rate counts: The AIRR Project. Join on ZOOM:https://umu.zoom.us/j/67206182274
Abstract: Pneumonia remains the leading cause of infectious childhood deaths globally, resulting in more annual deaths than HIV, TB and diarrhoea combined. Therefore, innovation in the diagnosis and treatment of pneumonia is needed to achieve Sustainable Development Goal 3.2. Standardised guidelines in low and middle-income countries (LMICs) currently include counting respiratory rate to diagnose paediatric pneumonia; however, this vital sign is often not collected or collected poorly, resulting in missed cases and sub-optimal treatment. We aim to develop and assess artificial intelligence (AI) methods for the analysis of mobile phone video data to determine respiratory rates in children under-five years. We will address this through three objectives: 1) develop and assess the diagnostic agreement of an AI algorithm for measuring respiratory rate and predictive signs of pneumonia from existing videos of children from LMICs, with and without signs of pneumonia. 2) Compare the AI accuracy against human rater accuracy. 3) Determine the appropriateness and acceptability of an automated video-based mHealth app for measuring respiratory rate amongst caregivers in South Africa.
