Health Data Science / 2021 / Article / Fig 4

Research Article

Probing Patient Messages Enhanced by Natural Language Processing: A Top-Down Message Corpus Analysis

Figure 4

Keyword frequencies per department. The top 30 most frequently used keywords per department were identified by MedTagger. Each row represents the keywords from a particular message type (i.e., A/P/L/U) per department. Large circles represent department types and are colored light orange (dermatology), light red (cardiology), and light green (gastroenterology or GIH). Small circles are sized according to the keyword frequency value and colored according to the medical concept of the keyword (i.e., ANAT/FIND/DISO/etc.), using the same color scheme as shown in Table 1.

Please note that the Table of Contents page is currently not displaying content published since June 7, 2021. We are working to fix this issue as soon as possible and apologize for the inconvenience.