Leveraging Machine Learning to Analyze Socioeconomic Disparities in U.S Cancer Clinical Trials

Cancer affects various segments of the U.S. population, but certain groups face a more significant burden because of social, environmental, and economic hardships. Using machine-learning techniques and the County-Level Socioeconomic and Cancer Clinical Trial Data, this study investigates the relationship between socioeconomic variables and the availability of cancer clinical trials in U.S. counties. Our research shows notable differences in trial availability according to socioeconomic parameters, including poverty rates, median income, and mortality rates. While the other methods demonstrated initial accuracies of 72.79%, 67.16%, and 62.25%, gradient boosting achieved approximately 75% prediction accuracy. These understandings can be used to focus therapies better, enhance healthcare equity, and enhance the results of cancer treatment.