Researchers Develop Novel Machine Learning Algorithm to Predict Length of Stay in TAVR Patients

Strategic Insights -

Researchers from the University of California San Francisco (UCSF) have made new findings in the field of artificial intelligence. According to their research, a novel machine learning (ML) algorithm was developed to predict early and late hospital discharge in patients who have undergone transcatheter aortic valve replacement (TAVR). The study analyzed data from 9,172 outpatient TAVR procedures conducted across 21 centers in the United States between 2017 and 2021. The algorithm identified variables involved in short and prolonged length of stay (LOS) following TAVR, which may facilitate targeted quality improvement programs to decrease LOS post-TAVR.

Key Takeaways:

  • Researchers from the University of California San Francisco (UCSF) developed a novel machine learning (ML) algorithm to predict early and late hospital discharge in patients who have undergone transcatheter aortic valve replacement (TAVR).
  • The study analyzed data from 9,172 outpatient TAVR procedures conducted across 21 centers in the United States between 2017 and 2021.
  • The algorithm identified variables involved in short and prolonged length of stay (LOS) following TAVR, which may facilitate targeted quality improvement programs to decrease LOS post-TAVR.
  • The study suggested that ML algorithms may have an important role in identifying novel predictors of short and prolonged LOS following TAVR.
  • The research included a multidisciplinary team of authors from the UCSF, including MD Gregory L. Judson, PhD Jeff Luck, BA Skye Lawrence, MPH Rakan Khaki, MD Harsh Agrawal, MD Krishan Soni, MD Kirsten Tolstrup, MD Vijayadithyan Jaganathan, and MD Vaikom S. Mahadevan.
  • The study was published in the journal JACC: Advances, a peer-reviewed publication accepted by Elsevier.

Statistics:

  • The study analyzed data from 9,172 outpatient TAVR procedures conducted across 21 centers in the United States between 2017 and 2021.
  • The algorithm identified variables involved in short and prolonged length of stay (LOS) following TAVR.
  • ML algorithms may have an important role in identifying novel predictors of short and prolonged LOS following TAVR.
  • The study suggested that targeted quality improvement programs can decrease LOS post-TAVR.

Sources:

  • Predictors of Length-of-Stay Among Transcatheter Aortic Valve Replacement Patients Using a Supervised Machine Learning Algorithm. JACC: Advances, 2025, 4(8): 101902.
  • University of California San Francisco (UCSF), San Francisco, California, United States.
  • NewsRx. University of California San Francisco (UCSF) Researchers Yield New Data on Machine Learning (Predictors of Length-of-Stay Among Transcatheter Aortic Valve Replacement Patients Using a Supervised Machine Learning Algorithm). Information Technology Newsweekly. August 12, 2025; p 984.