Page 73 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications

3

Application, algorithm, tools

directly related to deep learning

S. Shajun Nisha*, M. Mohamed Sathik, M. Nagoor Meeral
PG & Research Department of Computer Science, Sadakathullah Appa
College, Tirunelveli, Tamil Nadu, India
*Corresponding author: shajunnisha78@gmail.com

1. Introduction
Deep learning (DL) is at the pioneer of what machines can do,
and developers and business leaders absolutely need to use sense
of what it is and how it works. This unique type of algorithm has
far bettered any previous benchmarks for classification of various
images, text, and voice.
It also powers some of the most impressive applications in the
entire world, such as autonomous vehicles and real-time transla-
tion. There was certainly a knot of excitement around Google's
DL based in the world, but the business applications for this
eminent technology are more abrupt and potentially more effec-
tual [1]. The concept of Deep Learning is illustrated in Fig. 3.1 [1].
DL is a specific subcategory of machine Learning, which is also
a specific subset of artificial intelligence. For individual
definitions:
• Artificial intelligence is the broad edict of creating machines
that can think intelligently.
• Machine learning is one way of simplifying things, by using
various algorithms to glean insights from meta data.
• Deep learning is a way of doing by using a specific kind of
algorithm called a neural network.
Basic three types of scales that drive a DL process are data,
computation time, and algorithms. To improve the computation
time of the particular network, activation function plays an
important role. If sigmoid activation function is used, then graph
appears as shown in Fig. 3.2 [1].

Handbook of Deep Learning in Biomedical Engineering. https://doi.org/10.1016/B978-0-12-823014-5.00007-7
Copyright © 2021 Elsevier Inc. All rights reserved. 61