A randomized survival forest model developed via machine learning has been used to construct a prognostic model for cirrhosis associated with primary biliary cholangitis (PBC), according to a recent study published in the International Journal of Medical Sciences.
The model uses variables, such as white blood cell count and bile acid levels, to predict if a patient is likely to develop liver cirrhosis, an advanced stage of liver damage.
“There is an urgent need for more targeted treatment strategies to accurately identify high-risk patients. The use of survival forest models in machine learning is an innovative approach to constructing a prognostic model for PBC that can improve the prognosis by identifying high-risk patients for targeted treatment,” the researchers wrote.
The random survival forest model is a widely used machine learning algorithm that combines multiple variables into a single, more accurate prediction result.
Read more about PBC prognosis
This study collected and analyzed the clinical data and follow-up information of 90 patients with PBC-related cirrhosis diagnosed in Taizhou Hospital in China between January 2011 and December 2021. Data analyses and random survival forest model construction used the R programming language.
Next, the researchers conducted a one-way Cox regression analysis of 90 included samples and 46 variables. Seventeen variables with p-values below 0.1 were selected for initial model construction.
The final predictive model included five variables: cholinesterase, bile acid, the white blood cell count, total bilirubin, and albumin levels. All of these molecules and cells are important biomarkers in PBC. The model was used to accurately classify patients into high- and low-risk groups for developing liver cirrhosis.
