This study assesses the mechanism of influence and methods of improvement of big data auditing on the quality of internal control in hospitals against the background of smart healthcare construction. First, we build the framework of smart audit platform covering audit management, audit operation and audit collaboration system. Then, with the help of Lasso idea, the penalty parameters of the model are set and the variables that are not significantly related to the model are eliminated. After variable screening, the key variables of internal control quality of hospitals are retained by fusing the nonlinear model Logistic, and whether the internal control quality of hospitals meets the standard is assessed. The SHAP method was then utilized for feature attribution analysis to enhance the predictive efficacy and interpretability of the model. The results showed that the Lasso-Logistic model had over 90% accuracy in discriminating the quality of internal control in hospitals, revealing the non-linear association between the hospital’s large equipment utilization rate and prescription compliance rate with the quality of internal control. This study not only quantifies the impact of big data auditing on the quality of internal control in hospitals, but also provides a method for optimizing management strategies and improving the quality of internal control in hospitals.
