This study aimed to compare the qualitative and quantitative measurement of dorsal nigral hyperintensity (DNH) and its diagnostic efficacy for Parkinson's disease (PD) between 7 T MRI sequences: the combined liquid and blood vessel attenuation with refined double echo steady state (CLEAR-DESS) and susceptibility-weighted imaging (SWI). Following sequence simulation and optimization, a prospective study enrolled 33 PD patients and 52 healthy controls, all undergoing 7 T MRI with CLEAR-DESS and SWI. Four blinded observers qualitatively assessed image quality and DNH visibility, while two independent readers performed quantitative analyses of signal-to-noise (SNR) and contrast-to-noise ratios (CNR). Excellent inter-observer agreement was noted for both qualitative assessments. Diagnostic performance at the DNH level showed CLEAR-DESS (4 observers; AUC: 0.90-0.96; sensitivity: 85%-96%; specificity: 88%-94%) significantly outperformed SWI (4 observers; AUC: 0.73-0.85; sensitivity: 65%-70%; specificity: 74%-89%). Similarly, at the patient level, CLEAR-DESS (4 observers; AUC: 0.91-0.96; sensitivity: 91%-94%; specificity: 88%-100%) surpassed SWI (4 observers; AUC: 0.79-0.87; sensitivity: 80%-89%; specificity: 60%-79%). CLEAR-DESS also showed significantly better DNH visualization slices, SNR, and CNR than SWI (all p < 0.001). The 7T CLEAR-DESS sequence provides superior DNH visualization and PD diagnostic performance compared to SWI, indicating its potential as a preferred choice for DNH evaluation.