Low-quality ultrasonic medical images pose a significant challenge for the diagnosis and analysis of medical imagery, due to their inherent insufficient resolution and detail blurriness, especially in the precise identification and analysis of pathological sites. The deficiency in image quality can lead to difficulties for diagnostic personnel in accurately judging the condition of lesions, thereby adversely affecting the accuracy and efficiency of clinical diagnoses. To tackle this problem, this research introduces an enhanced progressive super-resolution reconstruction technique designed for pathological areas in breast cancer ultrasound imagery. This method employs a staged diffusion model framework, initially divided into a coarse prediction phase and a refinement phase. By iteratively refining the coarse prediction results in the refinement phase, the overall quality of medical images is significantly improved, making the images more precise in terms of clarity and detail representation, and thereby assisting medical professionals in diagnosing and analyzing lesions with higher accuracy. Experimental results from this study confirmed the efficacy and applicability of the proposed method in generating high-resolution medical images across various magnification levels.(2x and 4x super-resolution reconstruction), showcasing its significant potential in enhancing the quality of ultrasonic medical images.