IMAGING PROCESSING
Contents
List of figures
Figure 1: Medical Image of a bone to be processed. 5
Figure 2: Interpolated Image of a fractured bone. 6
Figure 3:Image filtering to improve resolution. 7
1.0 Introduction
Image processing is the procedure of performing manipulations on an image that is desired so that we can get an improved image and also obtain relevant data from it. Furthermore, image processing can be described as the use of computer processing systems or algorithms to make, process, and consequently present digital images. Image dispensation can be categorized as either being an analog or digital image processing. The analog method of processing is basically essential for the case of hard copies like prints and pictures. On the other hand, the digital image type of processing involves the application of computer software in order to manipulate the digital images accordingly to produce the desired outcome.
In our day to day activities, image processing is an integral aspect of our lives. This is because it forms part and parcel in an extensive range of fields such as science and technology and precisely in other areas such as robotics, remote sensing, medical diagnosis, and many more. Image processing finds its suitability in health care systems, this is attributed to the fact that it is pervasive, and the obtained result is very reliable. Image processing for medical diagnosis is vital as it can lessen the effect of noise, better the image, and improve its quality. The digitally processed medical images can precisely help focus on the disease and visually communicate medical and pathological information of the image.
This report will primarily focus on digital image processing using MATLAB simulation. MATLAB toolbox for medical image processing will be effectively used for medical image analysis in this report due to robustness as well as its ability to acquire, recognize, and process the images. Another useful feature with MATLAB is its ability in writing function files, or script files to perform the desired operations. The written scripts will be essential for documentation purposes, and these written files form a formal record for image processing used. They will guarantee that the final results can be openly tested and replicated by others should the need arise.
- Literature Review
In this section, a Literature review of digital image handling and processing and its application in the medical field by exploring its application using MATLAB software.
The earliest advancement and usage of image processing in the medical field go back into several decades, such as the sixties. The developed application method in the digital image processing fields was initially used in the jet engines. The application of image processing also used by the earliest researchers in satellite imaging, medical imaging, character recognition, and so forth. On the other hand, image processing its sometimes called picture processing. The picture processing was initially developed in the early sixties at the Massachusetts Institute of Technology in the Jet Propulsion Laboratory. Among the earliest researchers conducted were the application to satellite images, medical imaging, character recognition, and so forth. However, some researchers such as application in the satellite imaging, the use in wire photo conversion, use in medical imaging, and so on [1].
With the passing of time, image processing research advanced and the representation by Shanhui Sun Christian Bauer et al. [13] as an autonomous approach for differentiation of lung segments in the CT datasets for Medical purposes. The technique that is used was purposely developed and designed to clear sections to identify the lungs with cancerous masses. The exact steps that were followed mainly consisted of three essential steps. For instance, a ribcage detection algorithm was used to enhance an active kind or model of approach as used in the scheme of outlined by the section of the lungs. Lastly, the application matched the model and used it in the adapted as by using the optimal surface on finding approach.
Weixingwang et al. [16] also researched image processing, and he showcased the advancement in the application of the developed system to identify ridge detection algorithm to diagnose and identify the intermediate nodules appropriately. This mechanism is allowed to use main curative sections of malignant nodule levels. To avoid causing damages or cases of mortality of surgery. The application of the algorithm as compared to other methods such as traditional one in image segmentation algorithms. The outcome is the accurate results of the diagnosis.
Ashraf Anwar et al. [33] come up with cheap, easy to use, and multi-purpose image processing mechanism and data visualization program specifically outlined in MATLAB to find the brain tumors as early as possible. The use of this software to provide outstanding clinical and quantitative results on medical images. Some structural disparities led to brain disparities due to major and minor disorders as soon as possible. The use led to the discovery of disorders such as schizophrenia, epilepsy, and other brain-related disorders. The primary purpose of giving a particular diagnosis is to find the disease related to brain disorders and its psychic nature (Alzheimer’s disease). The application and use of medical imaging are quite expensive and very advanced in terms of use due to proprietary software and scarcity of experts.
Shirui Gao [35] focuses on the application of MATLAB, particularly in medical image processing mechanisms. The theoretical background and its uses. Through the use of MATLAB, this paper has focused on the current and post introduction of image processing capabilities. Furthermore, the application focuses on the medical fields through technology and image processing, which includes organ contours, results from interpretation, and filtering techniques. Finally, in DICOM image capabilities and the application of MATLAB widely.
The illustration by M Bister indicates the crucial points with a fast point being the implementation of bilinear interpolation, the use of watershed segmentation, and finally rendering using rendering with MATLAB. The application of MATLAB has been widely considered as an excellent environment for the application of the algorithm in development and generally used in medical routines. The image processing avail large database such as CT scan image provides hundreds of 512 by 512 slices. By adequately programming and considering vectorization, the specialization uses of MATLAB can run using C language
3.0 Application
The quality of an image is very vital health care systems due to the imaging aspects that are required to provide a dependable image of an organ or a bone so that accurate diagnosis and treatment can be administered. The application chosen for this report was a fractured bone.
4.0 MATLAB Simulation
In several years, the application of a medical imaging system increased considerably in dynamic progress. With time, there has been substantial improvement in various areas such as sensitivity, the level of resolution and acquisition speed. The development of new techniques has introduced and enhanced more development in processing of both analogue images and complimented by digital images. Therefore, the factors determining the quality of the image is developed. The quality of the image that has been acquired is processed by factors such as Campton scattering and photon attenuation. Also, the system parameters which involves both intrinsic and extrinsic of gamma camera system. The outcome of this factors is blurred and noisy image. The occurrence of blurred image in most time is the presence of immediate artefacts that results to fault diagnosis. Therefore, the images are demagnetized to the value of the physician and complement the follow up by specific processing.
Medical Image handing out techniques comprise all the conceivable tools used to change or analyze an image conferring to individuals’ needs. There are a number of graphical user interface tools used but, in this project, I used image tool (imtool) this allows one to interact with the image of interest that is to be processed.
Often some images may contain large data which may result to memory shortage, image may fail to load and slow in zooming. To enable handling of this data we use blockproc processing directory which facilitates high resolution processing of the images. The sequential steps followed in the project were as follows.
4.1 Image Reading to MATLAB
First use the directory (imread) function to access the data that is the medical image being processed which in my case was processing a fractured bone. >> x=imread(‘bones.png’);
>> imshow(x)
Figure 1: Medical Image of a bone to be processed
To analyze the fractured bone, the image tool (imtool) was used, with this tool a doctor can analyze the fracture and get to know the actual length of the crack before doing an operation. a particular region of interest is selected and the following techniques are carried out.
5.0 Results
5.1 Image Interpolation
This is done to resample the image and come up with a new one based on the existing one which enhances the image quality. In interpolation one is required to resize an image using (imresize).
z = imresize (Image, scale, method);
in this project we were able to find the length and width of the fracture.
>> x=imread(‘bones.png’);
>> imshow(x)
>> imtool(x)
Figure 2: Interpolated Image of a fractured bone
5.2 Contrast enhancement
Image contrast in image processing is the variance in strength equivalent to different focus of action in the patient. For high diagnostic precision, nuclear medical images must be of high distinction. This explains the distinction enhancement that was done for our simulation. Contrast enhancement was done to improve the visual clarity of the fracture bone using the command (imadjust) or new image=imcomplement(x);
5.3 Image filtering
The process of image filtering is applied for the purpose of removing noise and resolution enhancement and supplement for the loss of crucial details in the image processing by reducing noise. By carrying out filtering, the image with more resolutions and pixelated will be limited. The image processing and filtering it involves calculations of the image processing for both noise removal and recovery of resolution. The suppression of noise and enhancing image. Bearing all this in mind, the chances of faulty diagnosis will be low.
Figure 3:Image filtering to improve resolution
The filter mask consists of usually odd in size which is applied directly to the original data image. The mask is located at the center of each pixel where is the initial image. For each point of the mask, the values in pixel image undergo multiplication to the corresponding value of the mask.
The initial product of these outcomes are then added to sum up the value of the value into central pixel of the original image by being replaced. This procedure is repeated in every pixel in the image.
6.0 Conclusion
Image processing can be applied for health care systems images for diagnosis and treatment administration. The main importance attributed to this is the improvement in the attained image through extensive analysis of both qualitative and quantitative to gain important information data useful in patient’s therapy and care. With the application of MATLAB and Image Processing Toolbox, it is possible to provide both quantitative analysis and visualization of Medical images.
The objectives of carrying out this report were achieved. It was possible to carry out MATLAB simulation by ensuring the image processing steps were followed. The image that was used for the study was a fractured bone and all image processing steps were applied that produced image of greater quality and higher resolution which aided the diagnosis process.
7.0 References
Rosenfeld, A. (1976). Digital picture processing. Academic press.
Sun, S., Bauer, C., & Beichel, R. (2011). Automated 3-D segmentation of lungs with lung cancer in CT data using a novel robust active shape model approach. IEEE transactions on medical imaging, 31(2), 449-460.
Shankar, C. G., & Shanmugam, A. (2013). Compatible abnormality detection technique for CT and MRI brain images. The Imaging Science Journal, 61(7), 568-578.
Gao, S. (2013). Research on Medical Image Processing Method Based on the MATLAB. In Proceedings of the International Conference on Information Engineering and Applications (IEA) 2012 (pp. 269-276). Springer, London.
Bister, M., Yap, C. S., Ng, K. H., & Tok, C. H. (2007). Increasing the speed of medical image processing in MatLab®. Biomedical imaging and intervention journal, 3(1).