摘要
目的回顾性分析数字化乳腺摄影(FFDM)与磁共振检查在乳腺癌诊断中的应用价值。方法从2004年7月至2006年7月期间因乳腺肿块在我院接受数字化乳腺摄影及磁共振动态增强检查的患者共93例,与病理组织学对照,分析两种检查对乳腺癌的诊断能力以及乳腺癌的影像学表现特征。结果FFDM诊断乳腺癌敏感度84.48%,特异度80.00%,阳性预测值87.50%,阴性预测值75.68%,准确度为82.80%;多灶性病变检出率55.56%;FFDM簇状钙化灶检出率(29/69)明显高于MRI(7/69)。MRI诊断乳腺癌敏感度94.82%,特异度为97.14%,阳性预测值98.21%,阴性预测值91.89%,准确度为95.69%;多灶性病变检出率83.33%。病灶周围"毛刺征"、环形增强和快进快出型时间~信号曲线为乳腺癌特征性表现。结论FFDM对乳腺微小钙化灶敏感,适用于乳腺癌普查;MRI诊断乳腺癌敏感性、特异性高,是进行乳腺癌术前评价的有效方法。
Objective To retrospectively compare the efficacy of full-field digital mammography (FFDM) and magnetic resonance imaging (MRI) in the diagnosis of mammary cancer. Methods 93 cases suspected mammary lesions both received FFDM and MRI examinations . Compared with pathology, we analyzed the imaging features of mammary cancer in these two methods and investigated the capability of these two methods in mammary cancer diagnosis. Results The sensitivity, specificity, predictive value of positive cases, predictive value of negative cases and accuracy of FFDM in mammary cancer were 84.48%, 80.00%, 87.50%,75.68% and 82.80% respectively; the detection rate for multi-focus was 55.56%. And the detection rate for microcalcifications in FFDM examination (29/69) was much higher than in MRI (7/69). The sensitivity, specificity, predictive value of positive cases, predictive value of negative cases and accuracy of MRI in mammary cancer diagnosis were 94.82%, 97.14%, 98.21%, 91.89% and 95.69% ,respectively; the detection rate for multi-focus was 83.33%. The spiculate margin, ringed enhancement of the peripheral part of tumor and the fast-in-and-fast-out type of the time-signal intensity curve were the diagnostic signs of breast cancer. Conclusion FFDM was sensitive for breast microcalcifications and could be feasible for breast cancer screening. MRI had a high sensitivity and specificity for breast cancer diagnosis, which could be effective for pre-operation estimation.
出处
《影像诊断与介入放射学》
2007年第5期219-223,共5页
Diagnostic Imaging & Interventional Radiology
关键词
乳腺疾病
数字化乳腺摄影
磁共振成像
Breast neoplasms
Digital mammography
Magentic rescnance imaging