• Mukesh Kumar Saini
• Sachin Gupta
• Brett Fritze

Sepsis continues to be a leading cause of death and illness globally. Early detection algorithms may enhance patient outcomes, yet their real-world effectiveness is not widely studied. This research aimed to evaluate the impact of a deep-learning model, COMPOSER, on early sepsis prediction and patient outcomes. We conducted a quasi-experimental study across two Emergency Departments (EDs) within the Health Care System, analyzing data from 6,217 adult septic patients between January 1, 2016, and April 30, 2018. The intervention tested was a nurse-facing Best Practice Advisory (BPA) activated by COMPOSER. We compared metrics including in-hospital mortality, sepsis bundle adherence, changes in the Sequential Organ Failure Assessment (SOFA) score within 72 hours of sepsis onset, ICU-free days, and ICU encounters before and after the intervention. Using a Bayesian structural time-series analysis with confounder adjustments, we found that the implementation of COMPOSER was linked to a 1.9% absolute decrease (17% relative reduction) in in-hospital sepsis mortality (95% CI: 0.3%–3.5%), a 5.0% absolute increase (10% relative increase) in sepsis bundle compliance (95% CI: 2.4%–8.0%), and a 4% reduction (95% CI: 1.1%–7.1%) in SOFA score change within 72 hours post-sepsis onset. These results indicate that COMPOSER's early prediction capabilities were associated with significant improvements in mortality and sepsis bundle compliance.

Sepsis, Artificial Intelligence (AI), SERA Algorithm, Structured Data, Unstructured Data, Clinical Measures, Mortality Reduction, Healthcare Analytics, Machine Learning, Risk Assessment, Patient Outcomes

"Enhancing Quality of Care and Survival Outcomes with Deep Learning-Based Sepsis Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 11, page no.634-642, November-2018, Available :http://www.jetir.org/papers/JETIR1811d43.pdf

"Enhancing Quality of Care and Survival Outcomes with Deep Learning-Based Sepsis Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 11, page no. pp634-642, November-2018, Available at : http://www.jetir.org/papers/JETIR1811d43.pdf