Peripleural lung disease detection based on multi-slice CT images

Mikio Matsuhiro, Hidenobu Suzuki, Yoshiki Kawata, Noboru Niki, Yasutaka Nakano, Hironobu Ohmatsu, Masahiko Kusumoto, Takaaki Tsuchida, Kenji Eguchi, Masahiro KanekoMedical Imaging: Computer-Aided Diagnosis 2015 · OpenReview

Abstract

With the development of multi-slice CT technology, obtaining accurate 3D images of lung field in a short time become possible. To support that, a lot of image processing methods need to be developed. Detection peripleural lung disease is difficult due to its existence out of lung region, because lung extraction is often performed based on threshold processing. The proposed method uses thoracic inner region extracted by inner cavity of bone as well as air region, covers peripleural lung diseased cases such as lung nodule, calcification, pleural effusion and pleural plaque. We applied this method to 50 cases including 39 peripleural lung diseased cases. This method was able to detect 39 peripleural lung disease with 2.9 false positive per case.
BibTeX
@inproceedings{matsuhiro2015peripleural,
  author = {Mikio Matsuhiro and Hidenobu Suzuki and Yoshiki Kawata and Noboru Niki and Yasutaka Nakano and Hironobu Ohmatsu and Masahiko Kusumoto and Takaaki Tsuchida and Kenji Eguchi and Masahiro Kaneko},
  title = {{Peripleural lung disease detection based on multi-slice CT images}},
  booktitle = {Medical Imaging: Computer-Aided Diagnosis},
  year = {2015},
  url = {https://openreview.net/forum?id=QPoTLhzoR5},
}

Loading related content...

Figure