Image segmentation is the most important and first step in pattern recognition and image analysis. In this paper, an automatic segmentation algorithm based on light refraction was proposed for medical images. In the proposed algorithm, similarity percents of the pixels were calculated by using the amount of shift while occurred light through a transparent sheet and re-enters the same environment. The proposed algorithm is similar to region growing algorithm where the seed points are automatically selected and grown and does not require any prior knowledge of the number of regions existing in the image. So, it decreases the computational load required for the other image segmentation methods. The proposed algorithm is demonstrated by application to real medical images. Results have showed that proposed algorithm extract all segments effectively.