Image segmentation is a key step in medical image processing, since it affects the quality of the medical image in the follow-up steps. However, in the practice of processing MRI images, we find out that the segmentation process involves much difficulty due to the poorly defined boundaries of medical images, meanwhile, there are usually more than one target area. In this study, an improved algorithm based on the fuzzy connectedness framework for medical image is developed. The improved algorithm has involved an adaptive fuzzy connectedness segmentation combined with multiple seeds selection. Also, the algorithm can effectively overcome many problems when manual selection is used, such as the un-precise result of each target region segmented of the medical image and the difficulty of completion the segmentation when the areas are not connected. For testing the proposed method, some original real images, taken from a large hospital, were analyzed. The results have been evaluated with some rules, such as Dice's coefficient, over segmentation rate, and under segmentation rate. The results show that the proposed method has an ideal segmentation boundary on medical images, meanwhile, it has a low time cost. In conclusion, the proposed method is superior to the traditional fuzzy connectedness segmentation methods for medical images. basis of fuzzy connectedness, and achieved significant results in medical image segmentation [12].On the basis of the fuzzy connectedness framework, many scholars have put forward their own improved methods. On the whole, the improvements were mainly in the following two aspects:The first aspect was in the calculation method of fuzzy connectedness. Ciesielski et al. proposed an optimization calculation method of fuzzy connectedness, and applied it to medical image segmentation [13] and [14]. After that, they came out with an improved algorithm for multiple seeds, since the previous algorithm could only be applied to the single seed areas [15]. Jianjiang Pan et al. proposed an improved method in [16]. They pointed out that the use of native fuzzy connectedness calculation method would lead to poor segmentation effects on the images that have gradation of gray values. However, since their algorithm was based on a single seed pixel, the representation of the target area was insufficient. Also, in their algorithm, when the target area was long and thin, this would cause the target area to be incomplete.The second aspect was that the extensions of the fuzzy connectedness algorithm were mainly combined with other classic image processing methods. Bejar and Miranda combined the concept of fuzzy connectedness with the direction of region growing, excluding the results which were illogical [17]. Skoura et al. combined the fuzzy connectedness with feature detection of target area [18]. And Rueda et al. combined the fuzzy connectedness with the feature detection of the target region [19]. All these extensions enhanced the robustness of the classical fuzzy connectedness algorithm. Chunlan Yang et al. pr...
