Diabetic macular edema (DME) is one of the important reasons leading to blindness. Its pathological features are mainly manifested in the accumulation of fluid in the retina. A method for segmentation of diabetic macular edema in optical coherence tomography (OCT) retinal images is proposed. Firstly, through the image preprocessing, we exclude the impact of speckle noise and blood vessels on the final segmentation results. We used the improved level set method to solve the problem of segmentation effectively and calculated the area of edema area, which provides quantitative analytic tools for clinical diagnosis and therapy. Finally, we validated the method in this study on 15 OCT retina images with DME adults. The precision, sensitivity and dice similarity coefficient (DSC) for DME segmentation are 81.12%, 86.90% and 80.05%, respectively.