Page 63 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications

Chapter 2 Deep convolutional neural network in medical image processing 51

Figure 2.6 Application of CNN in medical imaging and healthcare, distribution of articles (A) year-wise and
(B) sources.

neural network in healthcare,”“CNN in medical imaging,” and
“convolution neural network in medical image processing.” It
can be analyzed that the application of CNN in the medical field
is increasing exponentially. Fig. 2.6B shows the distribution of ar-
ticles in different sources. It has been analyzed that 60% of total
work is published in journal articles.

6. Conclusion

The application of deep CNN has shown a huge scope in the
field of medical imaging in the past few years. In this regard,
the authors have tried to justify the use of deep CNN in the anal-
ysis of medical images. A brief summarization of different imag-
ing modalities is done in this chapter. Along with that, authors
have also given a broad description of the various medical imag-
ing processes such as segmentation, detection, and classification
of abnormalities, registration, and computer-aided detection or
diagnosis. As CNN has varied types of architectures, authors
have given a brief description of the different architectures that
are generally applied in medical imaging. As medical imaging is
being applied to different anatomical structures of the human
system, a detailed review of various works that have been done
using CNN is described individually for different organs. All the
studies are summarized in a table, which gives a general idea
about the work done along with the type of CNN architecture
used, the type of imaging modality used as well as the data set
that is used. Finally, a critical discussion is done on the future