目的:本研究旨在评估扩散峰度成像(DKI)和体素内不相干运动(IVIM)在前列腺癌(PCa)检测和表征中的诊断能力。
方法:对PubMed进行了全面搜索,Scopus,WebofScience,和Cochrane图书馆在2023年9月10日之前发表的文章,评估了MD的诊断功效,MK,Dt,f,和Dp参数。使用双变量混合效应回归模型汇总数据,并用R软件进行分析。
结果:总计,共纳入27项研究。分析揭示了DKI和IVIM的不同诊断功效。在整体模型中,敏感性和特异性分别为0.807和0.797,前瞻性研究显示更高的特异性(0.858,p=0.024)。检测模型提高了灵敏度(0.845)和特异性(0.812),DKI在两个指标中均优于IVIM(灵敏度:0.87,p=0.043;特异性:0.837,p=0.26);MD具有高灵敏度(0.88)和特异性(0.82),而MK的特异性显著较高(0.854,p=0.04);Dp的敏感性显著较低(0.64,p=0.016)。在表征中,敏感性和特异性分别为0.708和0.735,DKI和IVIM或Gleason评分之间没有显着差异;MK具有更高的灵敏度(0.78,p=0.039),f\的敏感性显著降低(0.51,p=0.019)。
结论:总之,该研究强调了DKI在检测PCa方面比IVIM具有更高的诊断准确性,MK因其精确度而脱颖而出。相反,诊断性能中的Dp和f滞后。尽管这些有希望的结果,该研究强调了标准化方案和研究设计的必要性,以实现可靠和一致的结果。
OBJECTIVE: This study aims to assess the diagnostic capabilities of Diffusion Kurtosis Imaging (DKI) and Intravoxel Incoherent Motion (IVIM) in prostate cancer (PCa) detection and characterization.
METHODS: A comprehensive search was conducted across PubMed, Scopus, Web of Science, and the Cochrane Library for articles published up to September 10, 2023, that evaluated the diagnostic efficacy of MD, MK, Dt, f, and Dp parameters. Data were pooled using a bivariate mixed-effects regression model and analyzed with R software.
RESULTS: In total, 27 studies were included. The analysis revealed distinct diagnostic efficacies for DKI and IVIM. In the overall model, sensitivity and specificity were 0.807 and 0.797, respectively, with prospective studies showing higher specificity (0.858, p = 0.024). The detection model yielded increased sensitivity (0.845) and specificity (0.812), with DKI outperforming IVIM in both metrics (sensitivity: 0.87, p = 0.043; specificity: 0.837, p = 0.26); MD had high sensitivity (0.88) and specificity (0.82), while MK\'s specificity was significantly higher (0.854, p = 0.04); Dp\'s sensitivity was significantly lower (0.64, p = 0.016). In characterization, sensitivity and specificity were 0.708 and 0.735, respectively, with no significant differences between DKI and IVIM or Gleason Scores; MK had higher sensitivity (0.78, p = 0.039), and f\'s sensitivity was significantly lower (0.51, p = 0.019).
CONCLUSIONS: In summary, the study underscores DKI\'s enhanced diagnostic accuracy over IVIM in detecting PCa, with MK standing out for its precision. Conversely, Dp and f lag in diagnostic performance. Despite these promising results, the study highlights the imperative for standardized protocols and study designs to achieve reliable and consistent outcomes.