UCLA researchers use machine learning on OCI to help predict surgical outcomes
The university taps Oracle Cloud Infrastructure to predict patient outcomes more accurately after surgery, resulting in better patient treatment.
“It would take over 30 hours for our standard machines to train an ensemble model for 2,000 patients. As our team tackled larger datasets of over 30,000 patients, Oracle Cloud provided significantly more computational power.”
The University of California, Los Angeles (UCLA) is a public university rooted in its land-grant mission of teaching, research, and public service. As part of its ongoing orthopedic research, UCLA uses the AutoPrognosis machine learning (ML) tool, which learns multiple ML models simultaneously and then automatically creates the best modeling pipelines for medical prognosis. However, UCLA lacked the computational resources needed to perform advanced ML analyses over a reasonable timeline. After the university migrated to Oracle Cloud Infrastructure (OCI), the team enjoyed high-performance computing and low-cost cloud storage options. OCI CPUs provide uniquely flexible and price-performant virtual machine and bare metal instances while on-demand block storage addresses UCLA’s storage workload requirements.