SUMMARY

• Incorporating artificial intelligence into sepsis is a complex and challenging issue.
• Artificial intelligence algorithms in sepsis may be applied to: prediction; prognosis; mortality prediction; and optimal management.

• Limited data suggests artificial intelligence may provide a new approach to sepsis, but optimal provider input and considerations remain elusive.

REVIEW

• Artificial Intelligence (AI) is the ability of computers to recognize patterns, inference and decision making.
• Although sepsis makes an ideal candidate for the application of AI, it is a complex and challenging problem due to:
  • Lack of a specific definition
  • Heterogeneity of various definitions
• There are several areas of sepsis identification and management where AI may be of benefit:
  • Early prediction of sepsis
  • Patient prognosis
  • Mortality prediction
  • Optimal management

• To date there are only 2 published prospective randomized clinical investigations assessing the potential benefit of AI sepsis models.

• These 2 studies demonstrate:

  • Improved antibiotic administration times from time of ED presentation

  • Improved hospital and ICU length of stay

  • Improved hospital mortality

CONCLUSIONS

• AI models may have differing outcomes due to significant heterogeneity of sepsis criteria.

• Area under the curve metrics should be combined with clinical outcomes

• The current AI models have limitations requiring further evaluation.

To receive articles like these in your Inbox, you can subscribe to Sepsis Program Optimization Insights.

Erkan Hassan is the Co-Founder & Chief Clinical Officer of Sepsis Program Optimization where he designs & oversees the implementation of solutions to optimize sepsis programs.

To discuss your organization’s Barriers of Effective Sepsis Care, contact Erkan by phone (844) 4SEPSIS (844-473-7747), email (erkan@spo.icu), or video chat.