AI Sheds Light on Predicting Post-Hip Fracture Mortality Risk

A model based on artificial intelligence, trained using fundamental blood and laboratory test data along with basic demographic information, was found to forecast a patient's likelihood of mortality within 1, 5, and 10 years after suffering a hip fracture. The findings are published in the Journal of Orthopedic Research . (1^✔ ✔Trusted Source
Machine Learning-Based Mortality Prediction in Hip Fracture Patients Using Biomarkers

Go to source)

1-Year Mortality Rates in Hip Fracture Patients Revealed

In the analysis of 3,751 hip fracture patient records from two in‐hospital database systems at the Beth Israel Deaconess Medical Center in Boston, the 1‐year mortality rate for all patients was 21% and for those aged 80 years and older was 29%. After assessing 10 different machine learning classification models, investigators found that the LightGBM model had the most accurate 1‐year mortality prediction performance.

Age, blood sugar levels, certain red blood cell characteristics, white blood cell levels, urea nitrogen levels, platelet count, calcium levels, and blood clotting time were factors with the highest predictive power. Most of these were also in the top 10 features of the LightGBM 5‐ and 10‐year mortality prediction models.

Reference:

1. Machine Learning-Based Mortality Prediction in Hip Fracture Patients Using Biomarkers - (https://onlinelibrary.wiley.com/doi/10.1002/jor.25675)

Source-Eurekalert