Recently Downloaded: Paper ID: ART20194595, Total 17 Articles Downloaded Today
Natural Language-Based Machine Learning Models for Information Extraction from Radiology Reports - Survey
Abstract: In the current digital world, most of the patients records are stored in the form of electronic health record, where vast amount of digital contents are reported based on the radiological information. The radiological reports are very important source about the patient and helps to researchers to improve the health care departments. The radiological reports are collected and saved for documentation and communication of image diagnosing. Since the radiological information are stored in a free text format, hence it requires appropriate automated information extraction to retrieve the structured data which helps the physician for decision making. Natural language processing (NLP) is the important technique that helps to attain the structured representation of radiological reports. The structured data further processed by the machine learning (ML) algorithm for classification purpose which helps the physician for better the decision making. In this review, the NLP and ML techniques are considered for handling the radiological reports. In this review, the list of approaches for the automatic classification of radiological reports are identified and gathered into four major ways which are rules based approach, machine learning based approach, and hybrid approach. Moreover, the drawbacks and the upcoming challenges, future scopes for enhancing the NLP functionality in radiology is described.
Keywords: Natural Language Processing, Electrical Health Records, Machine Learning, Classification, Radiology Reports.
Country: India, Subject Area: Computers in Biology and Medicine
Pages: 273 - 277
Edition: Volume 8 Issue 5, May 2019
How to Cite this Article?
Jewel Sengupta, "Natural Language-Based Machine Learning Models for Information Extraction from Radiology Reports - Survey", International Journal of Science and Research (IJSR), https://www.ijsr.net/archive/v8i5/show_abstract.php?id=ART20197568, Volume 8 Issue 5, May 2019, 273 - 277
Viewed 16 times.
Downloaded 8 times.