Machine Learning–Enhanced Frailty Assessment for High-Risk Older Patients Undergoing TAVR
2024

Machine Learning for Frailty Assessment in Older Patients

Sample size: 131 publication Evidence: moderate

Author Information

Author(s): Mardini Mamoun, Bai Chen, Price Catherine, Manini Todd, Al-Ani Mohammad

Primary Institution: University of Florida

Hypothesis

Can machine learning improve frailty assessments for older patients undergoing TAVR?

Conclusion

Using machine learning to combine clinical notes and structured data significantly improves frailty assessment in older patients.

Supporting Evidence

  • The study analyzed data from 131 TAVR patients collected over two years.
  • Machine learning models achieved an AUC of 0.82 when combining clinical notes and structured data.
  • Key frailty predictors included congestive heart failure management and back problems.

Takeaway

This study shows that using computers to analyze patient information can help doctors better understand how frail older patients are before heart surgery.

Methodology

The study used machine learning to analyze structured and unstructured data from clinical notes and electronic health records of TAVR patients.

Participant Demographics

Older patients undergoing TAVR.

Digital Object Identifier (DOI)

10.1093/geroni/igae098.0439

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication