Predictive model for abdominal liposuction volume in patients with obesity using machine learning in a longitudinal multi-center study in Korea
2024

Predicting Liposuction Volume in Obese Patients Using Machine Learning

Sample size: 9856 publication Evidence: high

Author Information

Author(s): Sang Hyunji, Park Jaeyu, Kim Soeun, Lee Myeongcheol, Lee Hojae, Lee Sun-Ho, Yon Dong Keon, Rhee Sang Youl

Primary Institution: Kyung Hee University Medical Center

Hypothesis

This study aimed to develop and validate a machine learning (ML)-based model for predicting liposuction volumes in patients with obesity.

Conclusion

The random forest regressor model best predicted the liposuction volume, suggesting that machine learning can enhance decision-making in surgical interventions.

Supporting Evidence

  • The random forest regressor model achieved the lowest mean absolute error (MAE) of 0.197.
  • Body fat mass and waist circumference were identified as the most important predictors of liposuction volume.
  • A web-based application was developed to suggest ideal liposuction volumes based on the predictive model.

Takeaway

Researchers created a smart computer program that helps doctors figure out how much fat to remove during liposuction for patients who are overweight.

Methodology

The study used longitudinal cohort data from five centers, applying various machine learning algorithms to predict liposuction volumes based on 15 patient-related variables.

Potential Biases

Selection bias may have influenced the findings as the dataset was derived from a single institution with patients who voluntarily underwent liposuction.

Limitations

The study's findings may not be generalizable due to reliance on data from a single South Korean institution and potential selection bias.

Participant Demographics

The study population consisted of 9,856 patients, primarily female (86.3%), with a mean age of 38 years and a mean BMI of 26 kg/m2.

Digital Object Identifier (DOI)

10.1038/s41598-024-79654-y

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication