Bietti crystalline corneoretinal dystrophy (BCD) is an autosomal recessive retinal degeneration caused by CYP4V2 mutations. Early clinical manifestations include night blindness, followed by crystalline deposits and progressive degeneration of the retinal pigment epithelium (RPE), ultimately resulting in irreversible vision loss. Currently, no effective therapies are available. Patient-derived induced pluripotent stem cells (iPSCs) were differentiated into RPE cells to establish a disease model, and pathogenic CYP4V2 mutations were targeted for correction using prime editing (PE). Different PE strategies and delivery systems were systematically compared to evaluate their therapeutic potential. Notably, a convolutional neural network (CNN)-based phenotypic analysis platform was developed to rapidly identify pathological phenotypes from bright-field images without the need for additional staining or immunolabeling, enhancing both the efficiency and accuracy of editing outcome assessment. The results indicate that the iPSC model combined with genome editing and the CNN-based phenotypic analysis platform provides a robust system for investigating BCD pathogenesis and screening targeted therapeutic strategies, and establishes a model applicable to gene-editing evaluation in inherited retinal diseases.