IEEE Consumer Electronics Magazine - January/February 2020 - 65

Grant 17K14694, Grant 17H02110, and Grant
26108009, in part by the Research Fund of the
Telecommunications Advancement Foundation,
Japan-China Scientific Cooperation Program
under Grant 6171101454, in part by the International Exchange Program of National Institute of
Information and Communications, and in part
by the Collaboration Project of National Institute
of Informatics.
Figure 5. Extraction result of a blood vessel.
(a) Original image. (b) Proposed method.

stage, the number of regions where the GGO
shadow was recognized correctly as GGO was
16, and the number judged as GGO by mistaking
a blood vessel was 52. From the above-mentioned information, the identification accuracies
are obtained as TP ¼ 100% and FP ¼ 11.158%.

& REFERENCES
1. M. Firmino, G. Angelo, H. Morais, M. R. Dantas, and
R. Valentim, "Computer-aided detection (CADe) and
diagnosis (CADx) system for lung cancer with
likelihood of malignancy," Biomed. Eng. Online,
vol. 15, no. 2, pp. 1-17, 2016.
2. Y. Yoshino

et al., "Automatic classification of lung

nodules on MDCT images with the temporal

CONCLUSION
A temporal subtraction image generation
architecture was discussed for enhancing GGO
shadows in the chest CT images. In this architecture, global matching, local matching, and 3D
elastic matching were performed to generate an
initial temporal subtraction image. For the initial
candidate regions of GGO, further region extraction was performed by supervoxel and graph
cuts. The density feature quantity and shape feature quantity were calculated for the extracted
region. To reduce artifacts, voxel matching was
applied to non-GGO candidate areas. The fullwidth at half-maximum (FWHM) and the ratio
of artifacts to lung volume were calculated as
an evaluation of temporal subtraction images.
It showed the effectiveness of the proposed
method with an FWHM of 28.3 and a rate of artifacts of 0.101. For future works, the region
extraction methods will be explored to generate
temporal subtraction images more precisely
with respect to GGO shadows.

subtraction technique," Int. J. Comput. Assisted
Radiol. Surg., vol. 12, no. 10, pp. 1789-1798, 2017.
3. A. Kano, K. Doi, H. MacMahon, D. D. Hassell, and
M. L. Giger, "Digital image subtraction of temporally
sequential chest images for detection of interval
change," Med. Phys., vol. 21, no. 3, pp. 453-461, 1994.
4. T. Aoki, N. Oda, Y. Yamashita, K. Yamamoto, and
Y. Korogi, "Usefulness of computerized method for lung
nodule detection on digital chest radiographs using
similar subtraction images from different patients,"
Eur. J. Radiol., vol. 81, no. 5, pp. 1062-1067, 2012.
5. [Online]. Available: http://www.haigan-info.com/,
Accessed on: Jan. 16, 2017.
6. Y. Boykov and V. Kolmogorov, "An experimental
comparison of min-cut/max-flow algorithms for energy
minimization in computer vision," IEEE Trans. Pattern
Anal. Mach. Intell., vol. 26, no. 9, pp. 1124-1137,
Sep. 2004.
7. H. Kobatake, "A convergence index filter for vector
fields and its application to medical image
processing," Electron. Commun. Japan III, Fundam.
Electron. Sci., vol. 89, no. 6, pp. 34-46, 2006.
8. S. Kido and H. Shouno, "Temporal subtracted images
of thoracic CT by use of displacements of ribs and

ACKNOWLEDGMENT
This work was supported in part by the Leading Initiative for Excellent Young Researcher
of MEXT-Japan under Grant 16809746, in part by
the Grants-in-Aid for Scientific Research of Japan
Society for the Promotion of Science under

January/February 2020

elastic object model," Med. Imag. Inf. Sci., vol. 24,
no. 4, pp. 126-130, 2007.
9. T. Tokisa et al., "A temporal subtraction technique for
3-D non-rigid wrapping method based on free form
deformation from thoracic MDCT image," IEICE Tech.
Rep., vol. 112, no. 271, pp. 11-16, 2012.

http://www.haigan-info.com/

IEEE Consumer Electronics Magazine - January/February 2020

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - January/February 2020