Advancements in internet technology have facilitated remote transmission and sharing of medical images, enhancing the efficiency of healthcare services while introducing security risks such as privacy breaches and data tampering. Therefore, this paper proposes a medical image enhancement protection scheme based on STAR-TEBLID-DCT. Firstly, the STAR-TEBLID-DCT algorithm was used to extract the key feature vectors from medical images. Then, combined with the perceptual hash algorithm to generate the feature sequence of medical images, in order to enhance security, we applied logistic chaotic mapping to perform chaos encryption on watermark images. Finally, we embed and extract the watermark with the help of zero watermark technology. Experimental results demonstrate that the proposed algorithm exhibits exceptional capabilities in watermark extraction and recovery under geometric attacks, effectively safeguarding the personal information of patients in medical images. Furthermore, the proposed scheme offers robustness and stealthiness, showing significant potential for development in the field of medical imaging.