Page 298 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications

Chapter 10 Deep neural network in medical image processing 289

5. Conclusion

The “black boxes” was the term used to describe deep
learning methods until very recently. It is sometimes not enough
to have a good prediction system, particularly in medicine where
accountability is critical and can have serious legislative conse-
quences. This system must also be in a certain way express itself.
Medicine is a branch of study which is very critical to every
society, which is the reason for a certain amount of restrictions
associated with the use of technologies, especially powerful tech-
nologies such as machine learning/deep learning; furthermore,
establishing the efficiency and reliability of such algorithms is
of paramount importance. Having been said that machine
learning is too important and powerful technology to disregard,
there could be far-reaching benefits of using deep learning in
medical image analysis, for example, giving a helping hand to
underqualified or underexposed doctors/technicians or to help
patients in remote location or under privileged countries. There
is still a lot of room for improvement in the present algorithms
before they are ready to be used in a practical application in
the medical field.

References
[1] K. Pulli, A. Baksheev, K. Kornyakov, V. Eruhimov, Realtime computer vision
with OpenCV, Queue 10 (4) (2012) 1e17, https://doi.org/10.1145/
2181796.2206309.
[2] K. Fukushima, Neocognitron: a self-organizing neural network model for a
mechanism of pattern recognition unaffected by shift in position, Biol.
Cybern. 36 (4) (1980) 193e202, https://doi.org/10.1007/BF00344251.
[3] S.C.B. Lo, H.P. Chan, J.S. Lin, H. Li, M.T. Freedman, S.K. Mun, Artificial
convolution neural network for medical image pattern recognition, Neural
Netw. 8 (7e8) (1995) 1201e1214, https://doi.org/10.1016/0893-6080(95)
00061-5.
[4] Y. Lecun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied
to document recognition, Proc. IEEE 86 (11) (1998) 2278e2324, https://
doi.org/10.1109/5.726791, https://linkinghub.elsevier.com/retrieve/pii/
S0006291X05005954. U, http://ieeexplore.ieee.org/document/726791/.
[5] C. Mead, M. Ismail, Analog VLSI Implementation of Neural Systems, The
Springer International Series in Engineering and Computer Science,
Springer, US, 2012 https://books.google.co.in/books?id¼oNjTBwAAQBAJ.
[6] A. Krizhevsky, I. Sutskever, G.E. Hinton, ImageNet classification with deep
convolutional neural networks, Adv. Neural Inf. Process. Syst. 2 (2012)
1097e1105.
[7] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, et al.,
ImageNet large scale visual recognition challenge, Int. J. Comput. Vis. 115
(3) (2015) 211e252, https://doi.org/10.1007/s11263-015-0816-y, arXiv:
1409.0575.