自动乳腺超声诊断系统（ABUS）结合计算机辅助诊断系统（CAD）前后对乳腺恶性病灶的诊断价值对比

目的 本研究旨在探讨应用自动乳腺超声诊断系统（ABUS）结合计算机辅助诊断系统（CAD）对于不同经验医师诊断乳腺恶性病灶的价值。 方法 收集行ABUS检查的乳腺病例1452例，结果均经病理或随访1年以上证实。比较6名医师（3名低年资医师和3名高年资医师）使用CAD系统前后的诊断敏感度、特异度、AUC及平均阅读时间。结果 1452例病例中，270例为恶性，共有282个恶性病灶，674例为良性，共有695个良性病灶，508例为阴性。应用CAD系统前，低年资与高年资医师诊断乳腺癌的敏感度分别为87%、93%，使用CAD后提高到94%、94%，低年资医师的诊断敏感度前后差异具有统计学意义（P<0.05），高年资医师差异无统计学意义（P＞0.05）。6名医师在使用CAD系统前后诊断特异度均略有下降，但差异均无统计学意义（P＞0.05）。低年资医师在使用CAD系统前后的诊断符合率有所提高，ROC曲线下面积由0.85提高到0.89，差异具有统计学意义（P<0.05）。而高年资医师组，虽然ROC曲线下面积由0.91提高到0.92，但差异不具有统计学意义（P＞0.05）。所有医师使用CAD后的平均阅读时间均有不同程度的延长，差异具有统计学意义（P<0.05）。结论 虽然使用CAD后的平均阅读时间有所延长，但在可接受范围内，借助ABUS-CAD的阅读模式能大大提高医生诊断的准确度和敏感度，对于低年资医师帮助更大。

Value of automated breast ultrasound system (ABUS) with and without computer-aided detection (CAD) diagnosis in detecting breast cancer

Objective The study aims to investigate the value of automated breast ultrasound system (ABUS) combined with computer-aided detection (CAD) for the diagnosis of malignant breast lesions by different experienced readers. Methods A total of 1452 cases assessed on ABUS were collected. All had either pathology diagnosis or follow-up for 1 years. Compare the diagnosis sensitivity, specificity, area under curve (AUC) and reading time of 6 readers (experienced and inexperienced) before and after using CAD system. Results Of the 1452 cases, 270 were malignant (282 lesions), 674 were benign (695 lesions), and 508 were negative. Before using CAD, the sensitivity of novice readers and experienced readers were 87% and 93%, respectively; which was 94% and 94% respectively after using CAD. Statistic difference (P<0.05) was existed between the novice readers before and after using CAD; whereas no statistic difference was existed in the experienced readers before and after using CAD (P>0.05). Diagnostic specificity in all six readers decreased slightly by the use of CAD, yet there was no statistical significance (P>0.05). Diagnostic accuracy, measured by AUC, for novices without CAD was 0.85, for novices with CAD was 0.89, with significant statistic differences (P<0.05). Although there was improvement of AUC after using CAD in experienced reader (0.91 vs 0.92), but no significant statistic difference was found (P>0.05). Reading time extended for all readers with CAD, the differences were statistically significant (P<0.05). Conclusions Although the reading time after the use of CAD has been extended, but ABUS-CAD can greatly improve the diagnostic accuracy of novice readers, which is significant for young doctors.