Experts Gather at SDSC to Discuss Using Data and AI to Improve Responses to Mass Casualty Events

Article Content

Last month, computer science and AI researchers, school safety managers, 911 program directors, service providers and policymakers gathered for the Predicting Mass Casualty Events from 911 Data workshop at the San Diego Supercomputer Center (SDSC) at UC San Diego. Led by SDSC’s Director of Spatial Information Systems Laboratory Ilya Zaslavsky, the meeting focused on how best to utilize new types of 911 data streams for early detection of mass casualty events. Don Reich from Public Safety Network Americas (PSNA) and Jon Whirledge from INdigital facilitated the workshop, bringing their extensive connections with state agencies and commercial companies enabling the 911 call infrastructure across the U.S.

“We discussed technical and organizational challenges and opportunities of using 911 data streams with advanced AI models to improve response to mass casualty events, especially on school campuses,” Zaslavsky said. “By finding patterns across multiple calls coming from a geographic location, and utilizing Next Generation 911 (NG911) standards, we lay the groundwork of a novel analytical and predictive framework for a better-designed, nationwide prototype early alert for 911 responders.”

Zaslavsky said that the two-day event included 25 participants who discussed specific needs for improved 911 data models for real-time event detection and shared perspectives on robust emergency notification and response. The workshop was a crucial step in fostering multi-disciplinary collaboration among government, academic, industry and non-profit sectors to explore innovative uses of 911 data in public safety.

One of the event’s sessions included a review of  analysis opportunities provided by NG911 compared to legacy analog-based 911 data, and early experiences with AI modeling of the 911 data streams. “We were pleased to share our current project with the community,” said Don Reich of PSNA, which provides 911 data analytics for public safety answering points throughout the U.S.

Another session at the event was a panel on school safety, where speakers articulated common challenges and insights learned from school shooting events. The panel was led by Tom Wheeler, Office of the Indiana Attorney General, and former Assistant Attorney General for Civil Rights, United States Department of Justice, who discussed School Safety: Lessons Learned with attendees. He was joined by speakers representing San Diego County District Attorney’s Office, San Diego County Office of Education, and San Diego Psychiatric Emergency Response Team. Through a series of examples and case studies, the panel examined the vital role of school resource officers, the importance of timely data sharing and collaboration between various entities involved in school safety, and the role of new technical approaches in improving situational awareness and enabling rapid response to emergencies.

Discussion sessions at the event focused on data acquisition and data sharing for 911 data, predictive modeling, ethical and privacy concerns related to emergency notification, the critical public safety infrastructure, and the importance of partnerships for public safety efforts.

“Our primary take-away from the event was a better understanding of the need for a collective effort to overcome public safety challenges plaguing schools nationwide and technical issues involved in the overhaul of the legacy 911 system to make it more effective and efficient,” Zaslavksy said. “As the next steps, we will build on the collaborations started at the workshop to advance predictive modeling and notification prototypes that have the potential to be widely deployed by our government and industrial partners.”

Funding for this conference was provided by the National Science Foundation (grant no. 2330460).