A study published on September 20, 2023 in the Journal of Orthopaedic Research concludes that using an artificial intelligence (AI) model trained on blood, lab test, and demographic data can accurately predict a patient’s risk of death within one, five, and ten years following a hip fracture!
The study analyzed 3,751 hip fracture patient records and found that the one-year mortality rate was 21% for all patients and 29% for patients of age 80 and older. From the ten (10) machine learning classification models reviewed, the LightGBM model provided the most accurate one-year mortality predictions, identifying certain biomarkers such as age, blood sugar levels, red and white blood cell characteristics, and calcium levels as the most significant factors in mortality prediction. The study’s lead author emphasized the immediate usefulness of these biomarkers in assessing the risk of poor outcomes following hip fractures. The study concluded that machine learning-based models can accurately and effectively predict mortality in hip fracture patients using the correct biomarkers and AI models.
The ainewsarticles.com article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
