AI Tool Aims to Predict Sepsis, Transforming Early Detection

UC San Diego's COMPOSER AI predicts sepsis early, offering a big impact globally.

By Sunil Sonkar 2 Min Read
2 Min Read
AI Tool Aims to Predict Sepsis, Transforming Early Detection

Scientists at the University of California – San Diego have created a AI tool called COMPOSER. This tool helps in predicting sepsis before someone shows any signs of being sick. Sepsis is a global problem and affects people in great number each year. This new AI tool could make a big difference.

The COMPOSER algorithm silently operates in the background, continuously monitoring over 150 real-time patient variables. These include lab results, vital signs, medications, demographics and medical history, beginning its surveillance as soon as a patient checks into the emergency department.

The scientists found that using the COMPOSER tool led to 17% fewer people passing away. If the algorithm identifies multiple risk factors linked to sepsis, it promptly notifies nursing staff through the hospital’s electronic health record. The hospital team works together with doctors to quickly make a plan to treat the person as soon as they see the first signs of sepsis.


Shamim Nemati, who helped with the study, said the AI system can see patterns that are hard for people to notice. If risk patterns can be attributed to other conditions with higher confidence, no alerts are generated.

COMPOSER became operational in December 2022 and is currently in use in various in-patient units at UC San Diego Health. Gabriel Wardi, a helper in the study and doctor at UC San Diego Health, liked the AI tool because it helps give quicker life-saving treatment.

This smart AI tool can help not only in one place but also in many countries where sepsis is a big problem. A 2022 study revealed that over half of the patients in Indian intensive care units suffer from sepsis. Using smart AI tools like COMPOSER can really help make sepsis less common all around the world.

Share This Article