文献速递乳腺US

深度学习在自动乳腺超声中的病灶检测效能及其影响因素:一项先导研究

The lesion detection efficacy of deep learning on automatic breast ultrasound and factors affecting its efficacy: a pilot study.
PubMed 文献 · Br J Radiol · 2021-12-15

本研究纳入2016年5月至2017年6月接受自动乳腺超声和手持超声检查的女性397例,分为训练组、测试组和对照组,优化卷积神经网络检测自动乳腺超声病灶。训练集688个病灶区域,按体积、病灶和患者计算的敏感度分别为93.8%、97.2%和100%,每体积1.9个假阳性;测试集247个病灶区域敏感度分别为92.7%、94.5%和96.5%,每体积2.4个假阳性;对照组900个体积每体积0.24个假阳性。病灶>1 cm³敏感度98%,≤1 cm³为87%(p<0.05)。

要点

  • 深度学习检测自动乳腺超声病灶的敏感度高、假阳性低,但效能受病灶大小影响。
  • 病灶>1 cm³敏感度98%,≤1 cm³为87%(p<0.05)。
  • 不同乳腺组织构成、病灶形态和回声模式间敏感度与假阳性相似。

核心结论

  • 深度学习对自动乳腺超声病灶检测具有高敏感度和低假阳性,但受病灶大小影响。

研究设计与样本

  • 先导研究,纳入2016年5月至2017年6月接受自动乳腺超声和手持超声检查的女性397例,分为训练组(乳腺癌163例、良性病变33例)、测试组(57例)和对照组(144例),优化卷积神经网络检测自动乳腺超声病灶。

诊断标准与关键数值

  • 训练集688个病灶区域,按体积、病灶和患者计算的敏感度分别为93.8%、97.2%和100%,每体积1.9个假阳性。
  • 测试集247个病灶区域,按体积、病灶和患者计算的敏感度分别为92.7%、94.5%和96.5%,每体积2.4个假阳性。
  • 对照组900个体积,每体积0.24个假阳性。
  • 病灶>1 cm³敏感度98%,≤1 cm³为87%(p<0.05)。
  • 不同乳腺组织构成敏感度与假阳性:均匀型97.5%、2.1,不均匀型93.6%、2.1;不同病灶形态:肿块型96.3%、2.1,非肿块型95.8%、2.0;不同回声模式:均匀型96.1%、2.1,不均匀型96.8%、2.1。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

OBJECTIVES: The aim of this study was to investigate the detection efficacy of deep learning (DL) for automatic breast ultrasound (ABUS) and factors affecting its efficacy.

METHODS: Females who underwent ABUS and handheld ultrasound from May 2016 to June 2017 (N = 397) were enrolled and divided into training (n = 163 patients with breast cancer and 33 with benign lesions), test (n = 57) and control (n = 144) groups. A convolutional neural network was optimized to detect lesions in ABUS. The sensitivity and false positives (FPs) were evaluated and compared for different breast tissue compositions, lesion sizes, morphologies and echo patterns.

RESULTS: In the training set, with 688 lesion regions (LRs), the network achieved sensitivities of 93.8%, 97.2% and 100%, based on volume, lesion and patient, respectively, with 1.9 FPs per volume. In the test group with 247 LRs, the sensitivities were 92.7%, 94.5% and 96.5%, respectively, with 2.4 FPs per volume. The control group, with 900 volumes, showed 0.24 FPs per volume. The sensitivity was 98% for lesions > 1 cm3, but 87% for those ≤1 cm3 (p < 0.05). Similar sensitivities and FPs were observed for different breast tissue compositions (homogeneous, 97.5%, 2.1; heterogeneous, 93.6%, 2.1), lesion morphologies (mass, 96.3%, 2.1; non-mass, 95.8%, 2.0) and echo patterns (homogeneous, 96.1%, 2.1; heterogeneous 96.8%, 2.1).

CONCLUSIONS: DL had high detection sensitivity with a low FP but was affected by lesion size.

ADVANCES IN KNOWLEDGE: DL is technically feasible for the automatic detection of lesions in ABUS.

来源:Br J Radiol,查看原文。版权归原作者所有。

引用

  1. Luo, Xu, Tang, et al. The lesion detection efficacy of deep learning on automatic breast ultrasound and factors affecting its efficacy: a pilot study. Br J Radiol. 2021. PMID: 34860574. doi:10.1259/bjr.20210438

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。