SUMMARY:

• Artificial Intelligence is improving early sepsis detection within clinical decision support algorithms.

• The effectiveness of these algorithms can vary based on a variety of factors including parameters evaluated, patient populations and time to predictions.

• Future efforts should focus on identification of high risk patients, escalation markers, risk score generation.

REVIEW:

• Artificial intelligence (AI) has allowed sepsis detection algorithms to shift from simple rule based engines to high performing machines on large amounts of data.

• The use of AI clinical decision support algorithms (AI CDS) has resulted in improved applications within clinical practice such as:

• • Improved coding efforts

• • Adherence to medical guidelines (evidence based medicine)Improved treatment choices with avoidance of errors and adverse events

• • Differential diagnosis considerations

• User adoption is critical for AI CDS success.

• However, adoption requires the efficient functionality of a number of considerations:

  • Consistent performance across various patient populations

• • Accuracy

  • Timeliness

  • Cost effectiveness

  • Clinical relevance

  • Technical robustness

• Any AI CDS must have the pre-requisite factors outlined in the table below:

• In addition, AI CDS sepsis algorithms should ideally be able to:

  • Predict key outcomes:

• • • Mortality

• • • Length of stay

• • • ICU admission

• • Identify patients at risk of deterioration and escalation levels of care needed

• • Risk stratify patients

• • Incorporate outputs into existing clinical workflows

• Sepsis AI CDS algorithms face a number of challenges in order to meet the above desired functionality:

• • Data protection & security

  • Integration into various workflows at differing centers and protocols

  • Often have limited validation data

  • Lack of effective data visualization

• Sepsis AI CDS algorithms currently available

CONCLUSIONS:

• AI CDS sepsis predictive algorithms move from a rules based engine to software as a medical device.

• The desired goal of these algorithms in sepsis is greater diagnostic accuracy and shorter time to sepsis identification.

• In the future, they should help identify high risk patients and discriminate between bacterial and viral sepsis.

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.