Case Study: UCSF achieves predictive insights into spinal cord injury recovery and optimal blood-pressure guidance with DataRobot

UCSF-BASIC uses DataRobot and Operating Room Data to Predict the Outcomes of Patients with Traumatic Spinal Cord Injuries

UCSF’s TRACK‑SCI team at the Brain and Spinal Cord Injury Center sought data‑driven guidance for acute blood‑pressure management during surgery and ICU care for traumatic spinal cord injury (SCI) patients. To analyze operating‑room time‑series data and predict recovery outcomes, UCSF partnered with DataRobot, using DataRobot’s Automated Feature Discovery and predictive analytics on intraoperative blood‑pressure and heart‑rate readings.

DataRobot helped the team build predictive models from five‑minute interval vitals and injury characteristics, automatically uncovering novel features that showed time spent in high blood‑pressure regimes strongly affects the likelihood of neurological improvement. These DataRobot‑driven insights narrow the optimal blood‑pressure window clinicians should target and are being used by UCSF to develop evidence‑based intraoperative and ICU guidelines for SCI care.

UCSF

Adam Ferguson

Associate Professor