Multimodal Medical Image Fusion and Quality Evaluation

  • Jinling Wei, Xiaomo Yu


Different sensors in medicine, different imaging technologies or inspection methods have their own advantages and disadvantages, and the information provided varies greatly. Medical images have different image modes, it can be divided into anatomical images and functional images specifically. Anatomical image include computed tomography image, magnetic resonance image, and ultrasonic image, it provides anatomical details of the morphology of human body. Usually, for a doctor's diagnosis, to obtain more image information from images, the single imaging technology cannot meet the needs of diagnosis, in many cases, a diagnosis can only be made if medical images provide comprehensive and rich information. Single mode medical image has limitations in local detail information description, which is not enough to provide sufficient information for doctors to obtain disease diagnosis. Therefore, multimodal medical image fusion is an important method to fuse the complementary features of different imaging methods into a fused image. This paper mainly discusses the medical image fusion method based on multiresolution technology. It focuses on the image fusion method based on wavelet transform. On the basis of analyzing the characteristics of existing image fusion quality evaluation methods, the combination of subjective and objective factors, establishing an image fusion quality evaluation method. This paper uses the above image fusion method, two groups of CT and MR images and one group of b-ultrasound and SPECT images were fused by experiment., good fusion effect is obtained. The experimental results showed that the algorithm can maintain a good consistency with subjective experimental results and is better than existing index. It is more suitable for multi-modal medical fusion image evaluation.