Page 210 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications

200 Chapter 7 Early detection and diagnosis using deep learning

Holding numerous new studies going around in the world
have shown the capabilities of advanced DL methods that
include dealing with the complex data [5,6] of image recognition
[7], text categorization [8], and much more.

2.1 Motivation for use of deep learning in
diagnostics
Earlier manual procedures in medicine were considered safe
for use; during the training periods of any employee in the
healthcare sector, they are used to prepare and maintain the
handwritten lab records. Knowing that advancement is needed
in at least this area, people already employed in this field started
finding ways to ease their work and increase their proficiency.
Advancement in this could bring improvement in the workflow
and will also improve patient's care as the thought behind
combining intelligent technologies with the medical sector.
From that point of time, advancements are made every single
day to improve patient's health and also working toward the
capability of the technologies now used to deal with new
medical issues such as new types of viruses, worms, bacteria,
and so on.
The papers in medical sectors were then replaced by
electronic medical record machines. When this technology got
introduced, everybody knew that if they want technology to excel
in healthcare sector, they will definitely need to enhance the elec-
tronic data that are provided to the doctors by incorporating the
powerful analytics and ML knowledge.
With the help of advanced analytics [9], doctors can have well-
maintained data and can access improved information related to
patient's care, not only data get improved but also the data
collected can be easily visualized, making it more presentable
and helping the doctors as well as patients to understand the
reports in a better way. Detailed data can be collected for vital
cryptograms of the patient making it very clear to the doctor
what kind of medications and treatment method should be pro-
vided to the patient, thereby decreasing the chance of life loss.
There is a requirement to develop more data to clinicians and
enable them to help them in taking improved pronouncements
when the patient is diagnosed with a certain disease and also dur-
ing the time of treatment selections, whereas being thoughtful of
the conceivable consequences and charge for separate one. The
worth of ML in healthcare sector is its capability to course through
the vast number of data sets, which are just beyond the possibility
of anthropoid competence, and thereby dependably translate