International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-12, October 2019
2429
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: L30111081219/2019©BEIESP
DOI: 10.35940/ijitee.L3011.1081219
Abstract: In this paper, the Osteoarthritis (OA) analysis in knee
radiographic images using artificial neural networks (ANN) is
considered. In Osteoarthritis, mobility is restricted and bones rub
each other causing extreme pain in knee due to cartilage
disintegration. The cartilage destruction is minimal in the initial
stage of OA. It is observed that a small number of researchers
have implemented identification and grading of Osteoarthritis
utilizing their own datasets for experimentation. However, there is
still need of automatic computer aided techniques to detect
Osteoarthritis for early recognition. In this work, a dataset of 1650
radiographic images of knee joints of OA patients are collected
from different hospitals and have been annotated by two different
orthopedic surgeons as per the Kellgren and Lawrence (KL)
grading system. To automate this grading procedure, the local
phase quantization and multi-block projection profile features are
computed from the images and then presented to artificial neural
network to classify the images based on the KL grading of the
severity of the disease. The classification accuracy of 98.7% and
98.2% with reference to surgeon-1 and surgeon-2 opinions,
respectively, is achieved.
Keywords: Knee Radiography, Osteoarthritis (OA), Local phase
quantization (LPQ), Multi-block Projection Profile (MB-PP),
Artificial Neural Network.
I. INTRODUCTION
In a knee anatomy, cartilage holds important role in leg
mobility. Cartilage is a bouncy material at the ends of the
bones that helps easy movement and abides as a shock
absorber. In Osteoarthritis (OA), cartilage is disintegrated
because of which bones rub each other causing extreme pain
and restricted mobility. The cartilage destruction is minimal
if the disease is diagnosed in the initial stage and analyzed
properly. The important clinical symptoms of OA in the
initial stage are joint pain in knee, hip, ankle, spine etc. If any
of these indications are experienced, the patients have to
immediately consult the doctor/experts preferably
Rheumatologists/Orthopedicians for diagnostic analysis. The
experts examine the patient clinically and may recommend
for a radiographic examination. Some of the important
radiological parameters are cartilage disintegration, reduced
joint space width, formation of osteophytes, loose bones and
bone deformation [8]. Depending on the radiological
Revised Manuscript Received on October 05, 2019.
Shivanand S.Gornale, Professor, Department of Computer Science,
School of Mathematics and Computing Sciences, Rani Channamma
University, Belagavi, Karnataka, India.
E-mail:shivanand_gornale@yahoo.com
Pooja U.Patravali, Research Scholar, Department of Computer Science,
School of Mathematics and Computing Sciences, Rani Channamma
University, Belagavi, Karnataka, India. E-mail: pcdongare@gmail.com
Prakash S.Hiremath, Professor, Dept. of Computer Science (MCA),
KLE Technological University, Hubballi-580031, Karnataka, India
Email:hiremathps53@gmail.com
parameters and severity level, the individual joint is
categorized into five different grades based on Kellgren and
Lawrence (KL) grading system [15] depicted in Table 1.
Table.1. Grading framework by Kellgren and Lawrence
[15]
KL Grades OA Analysis
Grade0 (Normal OA)
Radiographic parameters related to OA are
absent
Grade1 (Doubtful OA) Reduced joint space width
Grade2 (Mild OA)
Clear/ visible narrowing of joint space
Grade3 (Moderate OA) Numerous bony outgrowths, sclerosis
Grade4 (Severe OA)
Massive bone spurs, extreme sclerosis, bone
deformity
Generally, analysis of X-ray images is done manually by the
medical expert, which is time consuming, subjective and
sometimes unpredictable. The complexities related to the
medical images make it hard to examine them in an effective
way. Thus, to overcome these difficulties, an automated
method is proposed for early evaluation of Knee OA to
protect the cartilage and other tissues from the damage and
make the treatment more effective.
The rest of the paper comprises related work, proposed
methodology, experimental results and discussion along with
the summary.
II. RELATED WORK
Radiological parameters like joint space width, sclerosis and
osteophytes play very important role in assessment of
Osteoarthritis [15]. Based on these parameters, the medical
experts classify OA through manual inspection. From the
literature, it is found that numerous researchers have utilized
these radiological parameters for the detection of OA using
various machine learning and computer vision techniques.
Joseph Antony et al. [2] have used a method to localize knee
joint and classify the joint OA using fully convolutional
neural network and obtained better results. Jihye Lim et al.
[3] have used deep learning neural network for OA detection
using subjects‟ statistical and behavioral data. The result of
76.8% is achieved under the curve with scaled PCA. Yoo et
al. [1] developed a scoring system to predict radiographic
knee OA using KNHANES V-1 data and ANN. The results
attained were helpful in early prediction of Knee OA. Lior
Shamir et al. [4] have used WND-CHRM algorithm for the
early detection of Knee OA using computer aided analysis.
The classification rate of 91.5% for Moderate OA and 80.4%
for Minimal OA was achieved.
Shivanand Gornale et al. [9-14] have used contour based
segmentation method to detect the cartilage region using
Knee X-ray images [9-11].
Detection of Osteoarthritis in Knee Radiographic
Images using Artificial Neural Network
Shivanand S.Gornale, Pooja U. Patravali, Prakash S.Hiremath