Prediction of Head-Neck Cancer Recurrence from Pet/CT Images with Havrda-Charvat Entropy

Thibaud Brochet, Jerome Lapuyade-Lahorgue, Hua Li, Pierre Vera, Pierre Decazes, Su Ruan

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper proposes a loss function based on Havrda-Charvat entropy in deep neural networks for outcome prediction in head-neck cancers. Havrda-Charvat is a parameterized cross-entropy which generalizes the classical Shannon-based cross-entropy. Its parameter denoted α takes its values in ]0, ∞[ and one can recover some usual entropies, for instance Shannon for α = 1 or Gini coefficient for α = 2. In this paper, we propose to use this entropy to predict cancer recurrence by incorporating it in a neural network instead of Shannon's entropy for better adaptability. Our deep network is composed of a double auto-encoder to extract features and a classifier to predire cancer outcome. The experiments are conducted on MICCAI challenge dataset of Head-Neck cancer. The influence of the parameter on the results is studied and an optimal interval of its values is found. A result shows that Havrda-Charvat entropy can achieve better prediction performance than Shannon entropy, which is the most widely used in prediction task nowadays.

Original languageEnglish
Title of host publication2023 12th International Conference on Image Processing Theory, Tools and Applications, IPTA 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350325416
DOIs
StatePublished - 2023
Event12th International Conference on Image Processing Theory, Tools and Applications, IPTA 2023 - Paris, France
Duration: Oct 16 2023Oct 19 2023

Publication series

Name2023 12th International Conference on Image Processing Theory, Tools and Applications, IPTA 2023

Conference

Conference12th International Conference on Image Processing Theory, Tools and Applications, IPTA 2023
Country/TerritoryFrance
CityParis
Period10/16/2310/19/23

Keywords

  • CT images
  • Deep neural networks
  • generalized entropies
  • Havrda-Charvat entropy
  • head-neck cancer
  • parameter estimation
  • PET images
  • recurrence prediction
  • Shannon entropy

Fingerprint

Dive into the research topics of 'Prediction of Head-Neck Cancer Recurrence from Pet/CT Images with Havrda-Charvat Entropy'. Together they form a unique fingerprint.

Cite this