Result and Discussion - Segmentation of Hand Bone for Bone Age Assessment - page 122

Me an of quadruple scheme
Ex pe cted value of quadruple scheme
Me an of rigid scheme wi th row=4,column=4
Ex pe cted value of rigid scheme wi th row=4,column=4
Me an of rigid scheme wi th row=2,column=2
Ex pe cted value of rigid scheme wi th row=2,column=2
Me an of rigid scheme wi th row=3,column=3
Ex pe cted value of rigid scheme wi th row=3,column=3
Me an of rigid scheme wi th row=1,column=1
Ex pe cted value of rigid scheme wi th row=1,column=1

1.2

1

0.8

0.6

0.4

0.2

0

0 2 4 6 8 12 14 16 18

10

Ag e Group

Fig. 4.9 Comparisons of segmentation accuracy between manual rigid schemes with various
divisions and proposed automated fuzzy quadruple division scheme using the mean of FRAG and
expected value over all age group

input (in speciic context of this topic, the property referred to the size of block
radiograph) supplied to segmentation algorithm can be an effective alterna-
tive technique to improve the performance of segmentation algorithm instead of
implicitly improving the segmentation algorithm itself.

4.3.2.2 Evaluation on Quality Assurance Process

As shown in Fig. 4.6 f that even the diffused and equalized input block of hand
bone radiograph that have undergone ACR algorithm in adaptive quadruple
scheme failed to segment perfectly the desired hand bones; this is expected in
light of the mentioned challenges and difficulties that reside in the segmentation
techniques and the inherent properties of the radiographs itself and in fact, this
is the motivation of the proposed segmentation framework. However, thus far, it
is known that the most critical factor that influent the BAA, i.e., the number and
type of bones that determine the eventual bone age according to TW3 using RUS,
can be obtained via the previous processes from MBOBHE to the ACR algo-
rithm in quadruple division scheme. Therefore, the quality assurance process was
proposed and explained in last chapter that it functions as a filter that eliminates
the undesired areas and fills in the desired regions. Generally, it acts as artifacts
removers. This process is to enhance the segmentation accuracy automatically
with a few steps including entropy—based edge detection together with the pro-
posed BRANEA algorithm mentioned in last chapter. Therefore, in this chapter,