Sabottke Carl F, Spieler Bradley M
Department of Radiology, LSU Health Sciences Center New Orleans, 433 Bolivar St, New Orleans, LA 70112.
Radiol Artif Intell. 2020 Jan 22;2(1):e190015. doi: 10.1148/ryai.2019190015. eCollection 2020 Jan.
To examine variations of convolutional neural network (CNN) performance for multiple chest radiograph diagnoses and image resolutions.
This retrospective study examined CNN performance using the publicly available National Institutes of Health chest radiograph dataset comprising 112 120 chest radiographic images from 30 805 patients. The network architectures examined included ResNet34 and DenseNet121. Image resolutions ranging from 32 × 32 to 600 × 600 pixels were investigated. Network training paradigms used 80% of samples for training and 20% for validation. CNN performance was evaluated based on area under the receiver operating characteristic curve (AUC) and label accuracy. Binary output networks were trained separately for each label or diagnosis under consideration.
Maximum AUCs were achieved at image resolutions between 256 × 256 and 448 × 448 pixels for binary decision networks targeting emphysema, cardiomegaly, hernias, edema, effusions, atelectasis, masses, and nodules. When comparing performance between networks that utilize lower resolution (64 × 64 pixels) versus higher (320 × 320 pixels) resolution inputs, emphysema, cardiomegaly, hernia, and pulmonary nodule detection had the highest fractional improvements in AUC at higher image resolutions. Specifically, pulmonary nodule detection had an AUC performance ratio of 80.7% ± 1.5 (standard deviation) (0.689 of 0.854) whereas thoracic mass detection had an AUC ratio of 86.7% ± 1.2 (0.767 of 0.886) for these image resolutions.
Increasing image resolution for CNN training often has a trade-off with the maximum possible batch size, yet optimal selection of image resolution has the potential for further increasing neural network performance for various radiology-based machine learning tasks. Furthermore, identifying diagnosis-specific tasks that require relatively higher image resolution can potentially provide insight into the relative difficulty of identifying different radiology findings. © RSNA, 2020See also the commentary by Lakhani in this issue.
研究卷积神经网络（CNN）在多种胸部X光片诊断及图像分辨率下的性能变化。
本回顾性研究使用了公开可用的美国国立卫生研究院胸部X光片数据集，该数据集包含来自30805名患者的112120张胸部X光图像。所研究的网络架构包括ResNet34和DenseNet121。研究了从32×32到600×600像素的图像分辨率。网络训练范式使用80%的样本进行训练，20%用于验证。基于接收器操作特征曲线（AUC）下的面积和标签准确性评估CNN性能。针对每个考虑的标签或诊断分别训练二元输出网络。
对于针对肺气肿、心脏肥大、疝气、水肿、胸腔积液、肺不张、肿块和结节的二元决策网络，在256×256至448×448像素的图像分辨率下可实现最大AUC。在比较使用较低分辨率（64×64像素）与较高分辨率（320×320像素）输入的网络之间的性能时，在较高图像分辨率下，肺气肿、心脏肥大、疝气和肺结节检测的AUC分数改善最高。具体而言，对于这些图像分辨率，肺结节检测的AUC性能比为80.7%±1.5（标准差）（0.854中的0.689），而胸部肿块检测的AUC比为86.7%±1.2（0.886中的0.767）。
提高CNN训练的图像分辨率通常与最大可能的批量大小存在权衡，但最佳选择图像分辨率有可能进一步提高神经网络在各种基于放射学的机器学习任务中的性能。此外，识别需要相对较高图像分辨率的特定诊断任务可能有助于深入了解识别不同放射学发现的相对难度。©RSNA，2020另见本期Lakhani的评论。
