Multimodal Medical Image Fusion Based on Integer Wavelet Transform and Neuro-Fuzzy C. T. Kavitha 1 , C.Chellamuthu 2 1 ECE Department, Jerusalem College of Engineering, Pallikaranai, Tamil nadu, India. Email:ranjanakannan@yahoo.co.in 2 EEE Department, RMK Engineering College, Kavaraipettai, Tamil nadu, India. Email:-malgudi60@hotmail.com Abstract—Medical image fusion is used to derive useful information from multimodality medical image data. The idea is to improve the image content by fusing images like computer tomography (CT) and magnetic resonance imaging (MRI) images, so as to provide precise information to the doctor and clinical treatment planning system. This paper proposes image fusion based on Integer Wavelet Transform (IWT) and Neuro- Fuzzy. The anatomical and functional images are decomposed using Integer Wavelet Transform. The wavelet coefficients are then fused using neuro-fuzzy algorithm. Then Inverse Integer Wavelet Transform (IIWT) is applied to the fused coefficients to get the fused Image. The performance of this algorithm is compared with image fusion based on Discrete Wavelet Transform (DWT) and neuro-fuzzy using entropy metric. Fusion Symmetry (FS) which quantifies the relative distance in terms of mutual information of the fused image with respect to input images is measured. Fusion Factor (FF) the criterion of maximizing the joint mutual information is also quantified.  KeywordsŸMedical Image Fusion, Neuro-Fuzzy, Integer Wavelet Transform, Multimodality, CT, MRI, Entropy I. INTRODUCTION Image fusion refers to the techniques that integrate complementary information from multiple image sensor data such that the new images are more suitable for the purpose of human visual perception and computer processing tasks. The advantages of image fusion are improved reliability and capability ([2]-[4]). As the clinical use of various medical imaging systems extends, the fusion of multi-modality imaging plays an increasingly important role in medical imaging field. Different medical imaging techniques may provide scans with complementary and occasionally redundant information. The combination of medical images can often lead to additional clinical information not apparent in the separate images. However, it is difficult to simulate the surgical ability of image fusion when algorithms of image processing are piled up merely. So many solutions to medical diagnostic image fusion have been proposed today. Wavelet theory improves spatial resolution and spectral characteristics [1]. IWT not only maintains the characteristics of wavelet, but also has the features of fast operational speed and occupies less memory. This paper proposes image fusion using IWT and neuro-fuzzy. Registered medical CT and MRI images ([5], [6]) of the same people and same spatial part are used for fusion. The images are initially decomposed by Integer Wavelet Transform. The IWT coefficients are then fused by applying neuro-fuzzy fusion rule. Trimf membership function is used for fusing the coefficients. Then Inverse Integer Wavelet Transform is applied to the fused coefficients to get the fused Image. The performance of the fused image is calculated using entropy metric. Fusion Symmetry which quantifies the relative distance in terms of mutual information of the fused image with respect to input images is measured. Fusion Factor the criterion of maximizing the joint mutual information is also quantified. The rest of this paper is organized as follows. The integer wavelet theory is given in section II. Section III discusses about neuro-fuzzy based image fusion. The Proposed algorithm is given in section IV. Simulation results are given in section V. Section VI concludes the paper. II. INTEGER WAVELET THEORY Human visual system (HVS) is less sensitive to the high resolution detail bands like HL, LH and HH. Wavelet domain hides data in these regions. Hiding data in the high resolution regions increases the robustness while maintaining good visual quality. In discrete wavelet transform, the wavelet filters have floating point coefficients. When the data is hided in their coefficients, any truncations of the floating point values of the pixels that should be integers may cause the loss of the hidden information. It may lead to the failure of the data hiding system [10]. The problems of floating point precision of the wavelet filters is that when the input data is integer as in digital images, the output data will no longer be integer which doesn't allow perfect reconstruction of the input image [10]. Integer wavelet transform maps an integer data set into another integer data set. In IWT, there will be no loss of information through forward and inverse transform [11]. In the case of IWT, the LL subband appears to be a close copy of the original image, with smaller scale, while in the case of DWT, the resulting LL subband is distorted. Lifting schemes is used to perform integer wavelet transform. Lifting consists of three stages: split, predict and update [11]. The detailed steps are as follows. Split: The data set X i has some correlation structure and to obtain a more compact representation divide the original data set into odd subset {si 0 } and even subset { di 0 }. 296 978-1-4244-8594-9/10/$26.00 c 2010 IEEE