深度学习神经架构搜索鉴别乳腺锥光束CT良恶性病变
本研究利用神经架构搜索(NAS)自动生成卷积神经网络,在增强乳腺锥光束CT上鉴别良恶性病变,训练与验证来自一家机构,外部测试来自另一家机构。结果显示NAS生成的CNN模型在外部测试集AUC为0.727,敏感度80%、特异度60%,与最优ResNet-50及两位乳腺放射科医生的视觉评估无统计学差异。NAS可自动生成深度学习架构用于医学图像分析,无需专家设计。
要点
- NAS自动生成的CNN在增强BCBCT上鉴别乳腺良恶性病变性能与ResNet-50及放射科医生相当。
- 外部测试集敏感度80%、特异度60%,AUC 0.727。
- NAS方法无需专家设计网络架构,速度比ResNet-50快三倍。
局限性
- 样本量有限,为初步结果。
- 外部测试集特异度仅60%,假阳性率较高。
- NAS模型与放射科医生比较未报告具体读片一致性数据。
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
PURPOSE: To utilize a neural architecture search (NAS) approach to develop a convolutional neural network (CNN) method for distinguishing benign and malignant lesions on breast cone-beam CT (BCBCT).
METHOD: 165 patients with 114 malignant and 86 benign lesions were collected by two institutions from May 2012 to August 2014. The NAS method autonomously generated a CNN model using one institution's dataset for training (patients/lesions: 71/91) and validation (patients/lesions: 20/23). The model was externally tested on another institution's dataset (patients/lesions: 74/87), and its performance was compared with fine-tuned ResNet-50 models and two breast radiologists who independently read the lesions in the testing dataset without knowing lesion diagnosis.
RESULTS: The lesion diameters (mean ± SD) were 18.8 ± 12.9 mm, 22.7 ± 10.5 mm, and 20.0 ± 11.8 mm in the training, validation, and external testing set, respectively. Compared to the best ResNet-50 model, the NAS-generated CNN model performed three times faster and, in the external testing set, achieved a higher (though not statistically different) AUC, with sensitivity (95% CI) and specificity (95% CI) of 0.727, 80% (66-90%), and 60% (42-75%), respectively. Meanwhile, the performances of the NAS-generated CNN and the two radiologists' visual ratings were not statistically different.
CONCLUSIONS: Our preliminary results demonstrated that a CNN autonomously generated by NAS performed comparably to pre-trained ResNet models and radiologists in predicting malignant breast lesions on contrast-enhanced BCBCT. In comparison to ResNet, which must be designed by an expert, the NAS approach may be used to automatically generate a deep learning architecture for medical image analysis.
引用
- Ma J, He N, Yoon JH, et al. Distinguishing benign and malignant lesions on contrast-enhanced breast cone-beam CT with deep learning neural architecture search. Eur J Radiol. 2021. PMID: 34388626. doi:10.1016/j.ejrad.2021.109878
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。