en
Scientific article
Open access
English

Classification of Oncology Treatment Responses from French Radiology Reports with Supervised Machine Learning

Published inStudies in health technology and informatics, vol. 294, no. Challenges of Trustable AI and Added-Value on Health, p. 849-853
Publication date2022-05-25
Abstract

The present study shows first attempts to automatically classify oncology treatment responses on the basis of the textual conclusion sections of radiology reports according to the RECIST classification. After a robust and extended manual annotation of 543 conclusion sections (5-to-50-word long), and after the training of several machine learning techniques (from traditional machine learning to deep learning), the best results show an accuracy score of 0.90 for a two-class classification (non-progressive vs. progressive disease) and of 0.82 for a four-class classification (complete response, partial response, stable disease, progressive disease) both with Logistic Regression approach. Some innovative solutions are further suggested to improve these scores in the future.

eng
Keywords
  • RECIST
  • Automatic classification
  • Oncology
  • Treatment response
  • Machine Learning
  • Natural Language Processing
  • Radiography
  • Radiology
  • Research Report
  • Supervised Machine Learning
Funding
  • Swiss Personalized Health Network – SPHN -
Citation (ISO format)
GOLDMAN, Jean-Philippe et al. Classification of Oncology Treatment Responses from French Radiology Reports with Supervised Machine Learning. In: Studies in health technology and informatics, 2022, vol. 294, p. 849–853. doi: 10.3233/SHTI220605
Main files (1)
Article (Published version)
Identifiers
ISSN of the journal0926-9630
114views
27downloads

Technical informations

Creation06/07/2022 2:17:00 PM
First validation06/07/2022 2:17:00 PM
Update time03/16/2023 8:43:35 AM
Status update03/16/2023 8:43:34 AM
Last indexation05/06/2024 11:55:01 AM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack