Improvement and Innovation Showcase 38: Statewide Deep-Learning Models to Estimate Length of Stay

Alex Al-Saffar, Data Scientist, Southern Adelaide Local Health Network.
Hospital length of stay (LoS) of patients is a crucial factor for the effective planning and management of hospital resources. There is considerable interest in predicting the length of stay in order to improve patient care and increase service efficiency.
Alex has developed a tripartite deep learning model for statewide LoS Estimation at the time of patient admission. The model performance exceeds a model that unrealistically utilizes the DRG of the patient. The model is re-trained for every service of SA Health to provide LoS for all patients with zero input from any clinician.
Dr Alex completed his PhD on Data-driven techniques for biomedical electromagnetic imaging at The University of Queensland in 2021. He has a patent imaging neural network and is currently working with SALHN as data-scientist. Alex's interest span mathematics, statistics & programming and He's the author and maintainer of multiple Python packages.