Patient-specific fine-tuning of convolutional neural networks for follow-up lesion quantification

Mariëlle J A Jansen, Hugo J Kuijf (Corresponding author), Ashis K Dhara, Nick A Weaver, Geert Jan Biessels, Robin Strand, Josien P W Pluim

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Purpose: Convolutional neural network (CNN) methods have been proposed to quantify lesions in medical imaging. Commonly, more than one imaging examination is available for a patient, but the serial information in these images often remains unused. CNN-based methods have the potential to extract valuable information from previously acquired imaging to better quantify lesions on current imaging of the same patient. Approach: A pretrained CNN can be updated with a patient's previously acquired imaging: patient-specific fine-tuning (FT). In this work, we studied the improvement in performance of lesion quantification methods on magnetic resonance images after FT compared to a pretrained base CNN. We applied the method to two different approaches: the detection of liver metastases and the segmentation of brain white matter hyperintensities (WMH). Results: The patient-specific fine-tuned CNN has a better performance than the base CNN. For the liver metastases, the median true positive rate increases from 0.67 to 0.85. For the WMH segmentation, the mean Dice similarity coefficient increases from 0.82 to 0.87. Conclusions: We showed that patient-specific FT has the potential to improve the lesion quantification performance of general CNNs by exploiting a patient's previously acquired imaging.

Original languageEnglish
Article number064003
Pages (from-to)064003
Number of pages15
JournalJournal of Medical Imaging
Volume7
Issue number6
DOIs
Publication statusPublished - Nov 2020

Bibliographical note

© 2020 The Authors.

Keywords

  • convolutional neural network
  • magnetic resonance imaging
  • patient-specific

Fingerprint Dive into the research topics of 'Patient-specific fine-tuning of convolutional neural networks for follow-up lesion quantification'. Together they form a unique fingerprint.

Cite this