Académique Documents
Professionnel Documents
Culture Documents
I. I NTRODUCTION
The incidence of femur fractures increased significantly Max/Min Rule, Product Rule, are fused to design as com-
among people over 60 years old. Most of these orthopedic bining classifier [4], [5]. Fracture is detected by computing
and trauma surgical interventions need careful classification the pixels value of each images column or row in [6], where
in order to decide the position and type of implant to be a threshold is set, there would be a fracture if the value is
used intraoperative. Automatic recognition and classification less than the threshold.
of femur fractures plays an important role in quantitative
In the above mentioned investigations, most of them either
analysis.
only detect the fractures or design a general classifier, such
For the fracture clinical diagnosis is widely based on
as traverse fracture, open fracture, simple fracture and com-
computer-aided images, such as X-ray, CT, ultrasound and
minuted fracture. In this paper, a subtle pertrochanteric frac-
MRI images, many investigations on fracture detection and
ture classifier is proposed, which can distinguish five types
diagnosis are based on image processing. Morphological
of pertrochanteric fracture. The classification model is shown
gradient based image segmentation is used in [1] to detect
in Fig. 1, the original CT image is denoised and smoothed
radius bone fracture. Fracture detection is performed by a
by Gaussian filter, and then the distance regularization level
registered active shape model in [2], where the pelvic CT
set method is used to segment the region of interest (ROI).
images are preprocessed by adaptive windowing, bound-
After that, the Canny edge detection is used to detect the
ary tracing and wavelet transform. Stacked random forests
construction of femur and then the image is divided into
feature fusion is used in [3] to detect fracture in X-ray
several parts according to the definition of pertrochanteric
images. After edge and shape features are extracted from
fracture.
bone, multiple classifiers, such as Back Propagation Neural
Network, K-Nearest Neightbour, Support Vector Machine, For there usually are noises in the segmentation, filters
are used to remove the smaller noise regions and edges
*Research supported by the grant from National Natural Science by mistaken segmentation as shown in Fig. 2. A continues
Foundation of China (NSFC) ( No. 61273256, No. 61305033 and No. region would be considered as a outliers if its size is less
U1233103F01).
L. Wang and H. Cheng are with Center for Robotics, University of than 10 10, and then this region would be moved out. A
Electronic Science and Technology of China, Chengdu, China 611731. continues line would be moved out if its length is less than
(email: eewangling@uestc.edu.cn, hcheng@uestc.edu.cn). one forth of the width of each image. The pertrochanteric
H. Lan and Y. Zheng and K. Li are with Affiliated Hospital of
Chengdu University, Chengdu, China 610106. (email: 13714777@qq.com, fracture is classified by comparing the difference of each
183729860@qq.com, Likainan1961@126.com). part with reference normal images.
3852
here R and > 0 are coefficients, g is an edge indicator
1
function of image I, which defined as g = (1+GI 2 ) and G
3853
Fig. 5: The partials of images. The first row is the reference images of a normal femur, the second row is the test images
of a trochanter fracture, the third and the forth rows are the extracted edges from each part of the test images.
R EFERENCES
[1] B. Swathika, Anandhanarayanan.K, Baskaran.B, and Govindaraj.R,
Radius bone fracture detection using morphological gradient based
image segmentation technique, Int. J. of Comput. Sci. and Inform.
Fig. 6: Five classes fracture images. From the first image to Tech., vol. 6, no. 2, pp. 110, Feb 2015.
the last image, it is two parts to six parts fracture respectively. [2] J. Wu, P. Davuluri, K. R. Ward, C. Cockrell, R. Hobson, and K. Na-
jarian, Fracture detection in traumatic pelvic CT images, Int. J. of
Biomed. Imag., 2012.
TABLE I: The recognition rates [3] Y. Cao, H. Wang, M. Moradi, P. Prasanna, and T. Syeda-Mahmood,
Fracture detection in x-ray images through stacked random forests
Classes Two Three Four Five Six feature fusion, in IEEE Int. Symp. Biomed.l Imag. (ISBI), April 2015,
Data Number 9 6 7 14 2 pp. 801805.
[4] N. Umadevi and S. Geethalakshmi, Multiple classification system for
Recognition Rate 8/9 5/6 5/7 12/14 2/2 fracture detection in human bone x-ray images, in Int. Conf. Comput.
Commun. Net.g Technol. (ICCCNT), July 2012, pp. 18.
[5] V. L. F. Lum, W. K. Leow, Y. Chen, T. S. Howe, and M. A. Png,
Combining classifiers for bone fracture detection in x-ray images, in
The recognition rates of each classes are shown in Table IEEE Int. Conf. Image Process., vol. 1, Sept 2005, pp. 11491152.
I. The average recognition rate is 84.2%. There are some [6] L. Nascimento and M. Graca Ruano, Computer-aided bone fracture
identification based on ultrasound images, in IEEE Portuguese Meeting
estimation error for the greater or lesser trochanter fracture on Bioeng. (ENBENG), Feb 2015, pp. 16.
are not clearly even diagnosed by human eyes. For the [7] Z. Chen, K. Li, and Z. Zhang, A finite element analysis of six-segment
definition of the five types femoral pertrochanteric fractures classification of femur intertrochanteric fracture, China J. Orthop.
Trauma., vol. 17, no. 5, pp. 433437, May 2015.
is a novel subject, there is fewer methods to compare. [8] C. Li, C. Xu, C. Gui, and M. D. Fox, Distance regularized level
set evolution and its application to image segmentation, IEEE Trans.
Image Process., 2010.
VII. C ONCLUSION [9] J. Canny, A computational approach to edge detection, IEEE Trans.
Pattern Anal. Mach. Intell., vol. PAMI-8, no. 6, pp. 679698, Nov 1986.
This paper explore the pertrochanteric fracture classifica-
tion in 3D CT-images. A distance regularization level set
method is used to segment ROI from noisy background,
Canny edge detect operator is used to extract fracture edges,
and then by comparing the difference of edges between
reference normal images and testing images, the fracture is
classified according to the definition of pertrochanteric frac-
ture. Compared with existed works, the proposed method can
classify subtle fractures on pertrochanteric. The experimental
results illustrated the good performance of this method. In
future work, we will discuss the sparse feature representation
and supervised machine learning method on large data sets.
3854