Prediction of Patient Choice Tendency in Medical Decision-Making Based on Machine Learning Algorithm (Preprint)

Abstract

As an early branch of artificial intelligence technology, machine learning algorithms can effectively simulate human behavior by training on the data of the training set.

We used the machine learning algorithm decision tree(DT), k-nearest neighbors(KNN) and support vector machine(SVM) to predict patient choice tendencies in medical decision-making. We aimed to assist doctors to understand patient tendencies, to provide a reference for the formulation of decision-making schemes in clinical treatment. Ultimately, the cost of communication between doctors and patients is reduced, and the quality of medical decision-making can be improved.

We conducted first-hand questionnaires and collected the tendency data of 248 patients from third-level grade-A hospitals in China to predict patient medical decision-making tendency. We then selected 12 patient influencing factors and four types of choice tendencies in medical decision-making , namely treatment effect, treatment cost, treatment side effects, and treatment experience. Combined with the data characteristics of this study, we tested the accuracy of three machine learning algorithms decision tree, support vector machine and k-nearest neighbors in predicting patient selection tendency.

When the DT was used to predict patients' tendency, the accuracy of treatment effect, treatment cost, treatment side effect and treatment experience were 80.00%, 60.00%, 56.00%, and 60.00%, the KNN was 78.00%, 66.00%, 74.00% and 84.00%, and the SVM were 82.00%, 76.00%, 80.00% and 94.00%. It can be seen that the accuracy rate of the SVM model is the highest overall.

Among the three kinds of machine learning classification algorithms, SVM has the highest accuracy overall, and the influencing factors are more prominent. Therefore, the prediction results have certain reference values and guiding significance for doctors to formulate clinical treatment plans. The research results are helpful to promote the development and application of a patient-centered medical decision assistance system, to resolve the conflict of interests between doctors and patients and assist them to realize scientific decision-making.

None