From the Departments of Medical Imaging (J.J.J.G., S.D.V., I.S.) and IQ Health (M.J.M.B.), Radboud University Medical Center, Geert Grooteplein Zuid 10, 6525 GA Nijmegen, the Netherlands; Department of Radiology and Nuclear Medicine, Amsterdam University Medical Center, Amsterdam, the Netherlands (K.M.D.); Department of Radiology and Nuclear Medicine, Haga Teaching Hospital, Den Haag, the Netherlands (J.K.v.R.); Department of Radiology, Gelre Hospitals, Apeldoorn, the Netherlands (A.F.v.R.); Department of Radiology and Nuclear Medicine, Maastricht University Medical Center, Maastricht, the Netherlands (J.B.H.); Department of Radiology, Diakonessenhuis, Utrecht, the Netherlands (D.B.N.); Department of Radiology, Canisius Wilhelmina Hospital, Nijmegen, the Netherlands (L.E.M.D.); Department of Psychological and Brain Sciences, University of California Santa Barbara, Santa Barbara, Calif (C.K.A.); Department of Psychology, University of Nevada, Reno, Nev (M.A.W.); Dutch Expert Centre for Screening, Nijmegen, the Netherlands (M.J.M.B., I.S.); and Technical Medicine Center, University of Twente, Enschede, the Netherlands (I.S.).
Radiology. 2024 Oct;313(1):e240237. doi: 10.1148/radiol.240237.
Background Mammographic background characteristics may stimulate human visual adaptation, allowing radiologists to detect abnormalities more effectively. However, it is unclear whether density, or another image characteristic, drives visual adaptation. Purpose To investigate whether screening performance improves when screening mammography examinations are ordered for batch reading according to mammographic characteristics that may promote visual adaptation. Materials and Methods This retrospective multireader multicase study was performed with mammograms obtained between September 2016 and May 2019. The screening examinations, each consisting of four mammograms, were interpreted by 13 radiologists in three distinct orders: randomly, by increasing volumetric breast density (VBD), and based on a self-supervised learning (SSL) encoding (examinations automatically grouped as "looking similar"). An eye tracker recorded radiologists' eye movements during interpretation. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of random-ordered readings were compared with those of VBD- and SSL-ordered readings using mixed-model analysis of variance. Reading time, fixation metrics, and perceived density were compared using Wilcoxon signed-rank tests. Results Mammography examinations (75 with breast cancer, 75 without breast cancer) from 150 women (median age, 55 years [IQR, 50-63]) were read. The examinations ordered by increasing VBD versus randomly had an increased AUC (0.93 [95% CI: 0.91, 0.96] vs 0.92 [95% CI: 0.89, 0.95]; = .009), without evidence of a difference in specificity (89% [871 of 975] vs 86% [837 of 975], = .04) and sensitivity (both 81% [794 of 975 vs 788 of 975], = .78), and a reduced reading time (24.3 vs 27.9 seconds, < .001), fixation count (47 vs 52, < .001), and fixation time in malignant regions (3.7 vs 4.6 seconds, < .001). For SSL-ordered readings, there was no evidence of differences in AUC (0.92 [95% CI: 0.89, 0.95]; = .70), specificity (84% [820 of 975], = .37), sensitivity (80% [784 of 975], = .79), fixation count (54, = .05), or fixation time in malignant regions (4.6 seconds, > .99) compared with random-ordered readings. Reading times were significantly higher for SSL-ordered readings compared with random-ordered readings (28.4 seconds, = .02). Conclusion Screening mammography examinations ordered from low to high VBD improved screening performance while reducing reading and fixation times. © RSNA, 2024 See also the editorial by Grimm in this issue.
背景 乳腺摄影的背景特征可能会刺激人类视觉适应，从而使放射科医生更有效地检测到异常。然而，目前尚不清楚是密度还是其他图像特征驱动视觉适应。目的 研究当根据可能促进视觉适应的乳腺特征对筛查性乳房 X 光检查进行批量阅读时，是否会提高筛查性能。材料与方法 本研究为回顾性多读者多病例研究，纳入了 2016 年 9 月至 2019 年 5 月间获得的乳腺 X 光片。13 名放射科医生分别以三种不同的顺序对每个包含四张乳腺 X 光片的筛查性检查进行解读：随机顺序、按体积乳腺密度（VBD）递增顺序和基于自我监督学习（SSL）编码的顺序（检查自动分组为“看起来相似”）。眼动追踪器记录了放射科医生在解释过程中的眼球运动。使用混合模型方差分析比较随机顺序阅读、VBD 顺序阅读和 SSL 顺序阅读的受试者工作特征曲线下面积（AUC）、敏感度和特异度。使用 Wilcoxon 符号秩检验比较阅读时间、注视指标和感知密度。结果 共纳入 150 名女性（中位年龄为 55 岁[IQR，50-63]）的 75 例乳腺癌和 75 例非乳腺癌的乳腺 X 光片。与随机顺序阅读相比，按 VBD 递增顺序阅读的 AUC 更高（0.93 [95%CI：0.91，0.96] 比 0.92 [95%CI：0.89，0.95]； =.009），但特异度无差异（89% [871/975] 比 86% [837/975]， =.04）和敏感度无差异（均为 81% [794/975 比 788/975]， =.78），且阅读时间更短（24.3 秒比 27.9 秒， <.001）、注视计数更少（47 次比 52 次， <.001）、恶性区域注视时间更短（3.7 秒比 4.6 秒， <.001）。对于 SSL 顺序阅读，AUC 无差异（0.92 [95%CI：0.89，0.95]； =.70）、特异度无差异（84% [820/975]， =.37）、敏感度无差异（80% [784/975]， =.79）、注视计数无差异（54 次， =.05）或恶性区域注视时间无差异（4.6 秒， >.99）。与随机顺序阅读相比，SSL 顺序阅读的阅读时间明显更长（28.4 秒， =.02）。结论 按 VBD 由低至高的顺序对筛查性乳房 X 光检查进行排序，可提高筛查性能，同时减少阅读和注视时间。© 2024 RSNA
