# Introduction steoporosis is a disease characterized by low bone mass and deterioration of bone tissues, leading to bone fragility and, consequently, an increase in fracture risk [2][3][4][5][6]. Loss of bone mass occurs as part of the natural process of aging [4]. Bones naturally become thinner (called osteopenia) as one grows older, because existing bone is broken down faster than new bone is made. As this occurs, the bones lose minerals, heaviness (mass), and structure, making them weaker. With further bone loss, osteopenia develops into osteoporosis. The thicker bones are, the longer it takes to develop osteoporosis. Although osteoporosis can occur in men, it is most common in women older than age 65. Since osteoporosis can develop undetected for decades until a fracture occurs, early diagnosis is important [2]. A BMD test is the only way to diagnose osteoporosis and determine risk for future fracture [5]. The painless, noninvasive test measures bone density and helps determine whether medication is needed to help maintain bone mass, prevent further bone loss, and reduce fracture risk. Several different machines measure bone density. Central machines, such as the dual energy x-ray absorptiometry (DXA or DEXA), measure density in the hip, spine and total body. Peripheral machines, such as radiographic absorptiometry (RA) [1], peripheral dual energy x-ray absorptiometry (pDXA). Dual X-ray absorptiometry (DXA) uses two different Xray beams, due to which there is an increase in exposure to radiation causing harmful side effects. Special types of x-ray machines, are used. These machines are costlier. pDEXA also uses very low doses of radiation, and the results are faster than conventional DEXA measurements. pDEXA has limited usefulness (compared to DEXA) for monitoring the effect of medication used to treat osteoporosis [2]. The PDEXA equipment is much costlier than ordinary X-ray equipment. This method is not recommended for pregnant women because of the radiation exposure to the baby (fetus).Intensity slicing method is proposed to highlight the affected regions in osteoporotic x-ray images in order to assist the doctors in the identification of osteoporosis affected regions which cannot be visually identified. Author : RMK Engineering College, Kavaraipetttai Tamilnadu, India. E-mail : rsk.ece@rmkec.ac.in II. # Material and Method We have made use of X -rays of osteoporotic and non osteoporotic patients at different locations of the body collected from hospitals and processed it using image processing tools of MATLAB. This method for the detection of osteoporosis consists of two modules. # a) Intensity Slicing for Bone Structures This module is used to highlight pores and the regions that have low Bone Mineral Density, which could be of very great use to the doctors in identifying the affected areas which are otherwise not visually detectable. In this method the radiograph is loaded as bitmap image and converted into gray scale. Then the region of interest in gray scale image is chosen, in order to neglect the black background of the radiograph, the mean is computed for the region of interest omitting the pixel intensities that are less than a scale of 5. This ensures that there is no interference due to the black even if it is accidentally chosen in the region of interest. The histogram (Histogram plot the pixel intensities in x-axis and the corresponding number of pixels in the y-axis) of the image is computed and stored in a variable, and then the index value of the histogram that has the maximum number of pixels is noted. The size of the region of interest is stored in a variable. Let the index value of the histogram that has the maximum number of pixels be L. Then the pixels with the intensity lesser than the mean value and lie within the range L-10 & L-30 are displayed as red pixels that visually highlight the affected regions. This module is used to highlight striations within the bone structures that have very low calcium deposits, striations are caused due to drastic reduction of Bone Mineral Density. This method for detecting the will be of very great use to detect striations which are not visible to the naked eye. In this method the radiograph is loaded as bitmap image and converted into gray scale. Then the region of interest in gray scale image is chosen, in order to neglect the black background of the radiograph, the mean is computed for the region of interest omitting the pixel intensities that are less than a scale of 5. This ensures that there is no interference due to the black even if it is accidentally chosen in the region of interest. Then the region of interest is fully scanned for intensity variations, this can be done by monitoring the variation of intensities from pixel to pixel. If the variation of pixel intensity is over the threshold value then the pixel with lesser intensity is displayed as red pixel. Consider the following figures 2.6 & 2.7. If the threshold value is set as 5, then all the pixels which have intensities 5 greater than that of the previous pixel in the same row and pixels which have intensities 5 greater than that of the next pixel in the same row will be highlighted as shown in figure . III. # Results and Discussion # a) Results Of Intensity Slicing for Bones The results of intensity slicing for bone structures are displayed in the following page, figures 3.1, and 3.3,3.5 display the image of the osteoporotic bone and the corresponding outputs of intensity slicing method for bone structures are displayed in the figures 3.2, 3.4, and 3.6 respectively. By examining the osteoporotic images and their corresponding output for intensity slicing method for bone structures it would be clear that, the regions that have low calcium deposits are highlighted clearly .The pores within the bone structures which are hardly visible to the naked eye are clearly highlighted in the outputs The results of intensity slicing for tracing striations are displayed in the following page, figures 3.7,3.9,3.11 display the image of the osteoporotic bone with striations, the corresponding outputs of intensity slicing method for tracing striations are displayed in the figures 3.8,3.10,3.12 respectively. By examining the osteoporotic images and their corresponding output for intensity slicing method for tracing striations, it would be clear that, the regions having low calcium deposits are highlighted clearly .The striations within the bone structures which are hardly visible to the naked eye are clearly highlighted in the outputs . From the analysis of various radiographs using Intensity Slicing method produce accurate results. Thus using this method, detection of osteoporosis can be done with a minimum exposure to radiations and it can be of very great use to the doctors in the analysis of xray images for detecting osteoporosis. # OUTPUTS OF INTENSITY SLICING FOR VARIOUS IMAGES # OUTPUTS OF INTENSITY SLICING FOR VARIOUS IMAGES # Global Journal of Researches in Engineering ![From experimentation it was learnt that the range L-10 & L-30 covers the pixels with intensities which are more prominent in osteoporotic radiographs. The comparison between the histogram of osteoporotic and non osteoporotic radiographs clearly show the reason for the selection of the range in the figures 2.1. -2.4. & Flow Chart 2.5.](image-2.png "") 2122![Figure: 2.1](image-3.png "Figure: 2 . 1 Figure: 2 . 2 Figure") 26![Figure 2.6 : Pixels before highlighting Figure 2.7 : Pixels after highlighting Global Journal of Researches in Engineering](image-4.png "Figure 2 . 6 :") 3132![Figure : 3.1 Figure : 3.2 Figure : 3.3 Figure : 3.4 Global Journal of Researches in Engineering](image-5.png "Figure : 3 . 1 Figure : 3 . 2 Figure") 3![Figure 3.7 : Osteoporotic X-ray 1 FIgure 3.8 : Osteoporotic X-ray image 1 with striations highlighted](image-6.png "Figure 3 .") 3![Figure 3.11 : Osteoporotic X-ray 3 Figure 3.12 : Osteoporotic X-ray image 3 with striations highlighted.](image-7.png "VolumeFigure 3 .") © 2012 Global Journals Inc. (US) J May J ## May J * Fractal analysis of radiographs: Assessment of trabecular bone structure and prediction of elastic modulus and strength SMajumdar TLin TMLink JMillard PAugat XOuyang DNewitt RGould MKothari HKGenant Medical Physics 26 1999 * The Evaluation of Osteoporosis: Dual Energy X-ray Absorptiometry in Clinical Practise HWWahner IFogelman Martin Dunitz Ltd 1994 * Osteoporosis: Pathophysiology and Clinical Management By Eric S. Orwoll MengkoTr JTPramudito WSEAS Transaction on Computers 3 2004. 2003 Michael Bliziotes Edition: illustrated Published by Humana Press Texture analysis of Femur Radiographs for Osteoporosis Assessment * Osteoporosis: Diagnosis and Management PierreJMeunier 1998 Taylor & Francis * Current Topics in Osteoporosis Hong-WenDeng Yao-ZhongLiu 2005 World Scientific