Introduction - Segmentation of Hand Bone for Bone Age Assessment - page 9

3. Propose the utilization of anisotropic diffusion and automate the anisotropic
diffusion that able to smooth the irregular and uneven texture within anatomical
structures.
4. Instill the human cognitive ability availing of rule-based fuzzy inference system
to an adaptive division scheme to search the optimized size for applying the
central fundamental algorithm automatically.
5. Design an automated quality assurance scheme to assess the segmentation
image and determine the need to restore the lost detail, eliminate unwanted seg-
mented regions or remain as it is.
1.5 Research Scopes
A few scopes associated with database, methodology and result analysis as fol-
lowing are set to define the boundaries of the project and specify the task to be
accomplished.
1. Research focuses on designing and performing the proposed segmentation
framework on Radius, Ulna and Short Bones (RUS) of gray scale left hand
bone radiographs for computer-aided skeletal age scoring system (TW3).
2. Developments of all algorithms and graphs plotting in this topic are implemented
by using the MATLAB R2012a. The developed image processing algorithm are
focused on spatial domain processing instead of frequency domain as the trans-
formed representation is unsuitable in the context of hand bone segmentation
where the spatial location of the anatomical structure is of prime importance.
3. A total of 1,400 hand radiographs that are implemented in testing and analyses
are acquired from hand bone online database, http://ipilab.org , which comprises
of both genders in four populations which are Caucasian, African American,
Hispanic, and Asian, ages ranging between 0 and 18, collected from Children's
Hospital Los Angeles (CHLA).
4. To have fair comparisons, the performance evaluations of proposed automated
segmentation frameworks are conducted in two methods:
a. Analytical evaluations method is performed on existing methods that are
non-automated where direct implementations without human interventions
are not possible or are subjective to justify the performance fairly.
b. Empirical evaluations method is performed on existing methods that are
automated where direct implementations are possible and objective to justify
the performance.
1.6 Research Contributions
The contribution of this topic involves addressing the problems of hand bone
segmentation followed by filling in the technical, theoretical and conceptual gap
between the basic segmentation methods and the state-of-the-art deformable