• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar
  • Skip to footer

Centre for Artificial Intelligence and Cybersecurity – AIRI

  • Home
  • About Us
    • Vision, Mission and Goals
    • Center Activities
    • Center Faculty
    • Steering Committee
    • Press
  • Research
    • Scientific Projects
    • Research Papers
  • Laboratories
    • Machine Learning
    • Natural Speech & Language Processing
    • Blockchain Technology
    • Information Processing & Pattern Recognition
    • AI in Medicine
    • Data Mining
    • Computer Vision
    • Human-Computer Interaction
    • Maritime Cybersecurity
    • Autonomous Navigation
    • AI in Mechatronics
    • AI in Education
    • Hybrid Computational Methods
    • Drug Design
    • Legal Aspects of AI
    • Ethically Aligned AI
    • Cultural Complexity
  • Collaboration
    • Industry Collaboration
    • Industry Projects
    • International Collaboration
  • News
  • Contact

Automatic Annotation of Narrative Radiology Reports

01.04.2020

Narrative texts in electronic health records can be efficiently utilized for building decision support systems in the clinic, only if they are correctly interpreted automatically in accordance with a specified standard. This paper tackles the problem of developing an automated method of labeling free-form radiology reports, as a precursor for building query-capable report databases in hospitals. The analyzed dataset consists of 1295 radiology reports concerning the condition of a knee, retrospectively gathered at the Clinical Hospital Centre Rijeka, Croatia. Reports were manually labeled with one or more labels from a set of 10 most commonly occurring clinical conditions. After primary preprocessing of the texts, two sets of text classification methods were compared: (1) traditional classification models—Naive Bayes (NB), Logistic Regression (LR), Support Vector Machine (SVM), and Random Forests (RF)—coupled with Bag-of-Words (BoW) features (i.e., symbolic text representation) and (2) Convolutional Neural Network (CNN) coupled with dense word vectors (i.e., word embeddings as a semantic text representation) as input features. We resorted to nested 10-fold cross-validation to evaluate the performance of competing methods using accuracy, precision, recall, and F1 score. The CNN with semantic word representations as input yielded the overall best performance, having a micro-averaged F1 score of 86.7%. The CNN classifier yielded particularly encouraging results for the most represented conditions: degenerative disease (95.9%), arthrosis (93.3%), and injury (89.2%). As a data-hungry deep learning model, the CNN, however, performed notably worse than the competing models on underrepresented classes with fewer training instances such as multicausal disease or metabolic disease. LR, RF, and SVM performed comparably well, with the obtained micro-averaged F1 scores of 84.6%, 82.2%, and 82.1%, respectively.

Authors:
Ivan Krsnik, Goran Glavaš, Marina Krsnik, Damir Miletić, Ivan Štajduhar
Journal:
Diagnostics
Publishing date:
01.04.2020
View original article

Primary Sidebar

Latest Projects

Machine Learning for Knowledge Transfer in Medical Radiology

Estimating River Discharges in Highly Stratified Estuaries

Multilayer Framework for the Information Spreading Characterization in Social Media during the COVID-19 Crisis (InfoCoV)

European Network for assuring food integrity using non-destructive spectral sensors

National Competence Centres in the Framework of EuroHPC (EUROCC)

Latest Research Papers

Rule-Based EEG Classifier Utilizing Local Entropy of Time–Frequency Distributions

Rethinking Effects of Innovation in Competition In The Era of New Digital Technologies

A Comparison of Approaches for Measuring the Semantic Similarity of Short Texts Based on Word Embeddings

Indoor Localization Based on Infrared Angle of Arrival Sensor Network

Gravitational-Wave Burst Signals Denoising Based on the Adaptive Modification of Intersection of Confidence Intervals Rule

Latest News

U Rijeci radi Centar za umjetnu inteligenciju, već su u prvoj godini rada povukli šest milijuna u 14 projekata

UNIRI Excellence Awards in Science

Talk on conference “Exploring Digital Legal Landscapes”

ICAIH 2020 conference presentation

International conference “Exploring Digital Legal Landscapes” – 11th of December, 2020

We provide the expertise for solving real world problems using AI

If your company wants to implement artificial intelligence in your products or services, or increase your level of cybersecurity, our multidisciplinary team of scientists is your ideal partner.

Contact us

Footer

Center for Artificial Intelligence and Cybersecurity
  • jlerga@airi.uniri.hr
  • +385 51 406 500

University of Rijeka

University of Rijeka

About the Center

  • About Us
  • News
  • Privacy Policy
  • Contact

Center Activities

  • Laboratories
  • Scientific Projects
  • Industry Projects
  • Research Papers
  • Industry Collaboration
  • International Collaboration

Footer bottom left

© 2020 Center for Artificial Intelligence and Cybersecurity, all rights reserved.

Designed & developed by Nela Dunato Art & Design