As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
In this paper, we propose a data mining method for exploring the decision-making processes of physicians from electronic patient records and test it on the medical records of patients with type-2 diabetes mellitus. This method runs in two modes: general and partitioned. In the general mode, it mines rules from the whole medical records. In the partitioned mode, with a given partition factor, medical records are assigned into categories and a corresponding set of rules will be discovered for each category. Medication prescription predictions can be provided based on these rules. By comparing mined rules and prescription prediction accuracy under different modes, we discover that: 1) both the averaged precision and recall rate of the general mode can reach around 80%; 2) physicians tend to conform to the guideline instead of having their own preferences; 3) the medication decision can be affected by some hidden factors. These findings suggest this method show promise in discovering physician practice patterns and obtaining insights from real medical records.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.