University of Twente Student Theses

Login

Automatic Registration of Clinical Auditsfor Head and Neck Oncology at MST

Kortstra, Wybren (2020) Automatic Registration of Clinical Auditsfor Head and Neck Oncology at MST.

[img] PDF
1MB
Abstract:Clinical audits are used to analyze the quality of health care and improve treatment. \ac{DHNA} is an audit form that contains around 180 items. The registration form is currently filled manually which is a time consuming process. Most data in \ac{EHR} systems is stored as plain text. This research aims to automate value extraction for items in this audit form. We proposed a solution that uses natural language processing to analyze the plain text and extract the There are two types of items that need to be registered: categorical and continuous. For categorical items we proposed classification methods that use medical text documents. We used different types of preprocessing to zoom in on relevant data to improve the classification results. For the continuous items we proposed a technique which adds labels to words in medical text documents. We found that classification without preprocessing scores higher than classification with the preprocessing, but when looking at the features that are most important to this score we found no relevant features. The labeling technique performed very well on the text and extracting the values for the continuous items was very successful as a result of that. Even though the methods for extracting information from the EHR are not perfect they can aid doctors in registering the patient information for DHNA.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science MSc (60300)
Link to this item:https://purl.utwente.nl/essays/80461
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page