Usefulness of fine-tuning for deep learning based multi-organ regions segmentation method from non-contrast CT volumes using small training dataset

Yuichiro Hayashi, Chen Shen, Holger R. Roth, Masahiro Oda, Kazunari Misawa, Masahiro Jinzaki, Masahiro Hashimoto, Kanako K. Kumamaru, Shigeki Aoki, Kensaku Mori

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents segmentation of multiple organ regions from non-contrast CT volume based on deep learning. Also, we report usefulness of fine-tuning using a small number of training data for multi-organ regions segmentation. In medical image analysis system, it is vital to recognize patient specific anatomical structures in medical images such as CT volumes. We have studied on a multi-organ regions segmentation method from contrast-enhanced abdominal CT volume using 3D U-Net. Since non-contrast CT volumes are also usually used in the medical field, segmentation of multi-organ regions from non-contrast CT volume is also important for the medical image analysis system. In this study, we extract multi-organ regions from non-contrast CT volume using 3D U-Net and a small number of training data. We perform fine-tuning from a pre-trained model obtained from the previous studies. The pre-trained 3D U-Net model is trained by a large number of contrast enhanced CT volumes. Then, fine-tuning is performed using a small number of non-contrast CT volumes. The experimental results showed that the fine-tuned 3D U-Net model could extract multi-organ regions from non-contrast CT volume. The proposed training scheme using fine-tuning is useful for segmenting multi-organ regions using a small number of training data.

Original languageEnglish
Title of host publicationMedical Imaging 2020
Subtitle of host publicationComputer-Aided Diagnosis
EditorsHorst K. Hahn, Maciej A. Mazurowski
PublisherSPIE
ISBN (Electronic)9781510633957
DOIs
Publication statusPublished - 2020
EventMedical Imaging 2020: Computer-Aided Diagnosis - Houston, United States
Duration: 2020 Feb 162020 Feb 19

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume11314
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2020: Computer-Aided Diagnosis
CountryUnited States
CityHouston
Period20/2/1620/2/19

Keywords

  • ct
  • deep learning
  • ne-tuning
  • segmentaion

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Usefulness of fine-tuning for deep learning based multi-organ regions segmentation method from non-contrast CT volumes using small training dataset'. Together they form a unique fingerprint.

  • Cite this

    Hayashi, Y., Shen, C., Roth, H. R., Oda, M., Misawa, K., Jinzaki, M., Hashimoto, M., Kumamaru, K. K., Aoki, S., & Mori, K. (2020). Usefulness of fine-tuning for deep learning based multi-organ regions segmentation method from non-contrast CT volumes using small training dataset. In H. K. Hahn, & M. A. Mazurowski (Eds.), Medical Imaging 2020: Computer-Aided Diagnosis [113143V] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 11314). SPIE. https://doi.org/10.1117/12.2551022