School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK, USA.
IBM, IBM Cloud, Armonk, New York, USA.
J Ultrasound Med. 2023 Jan;42(1):71-79. doi: 10.1002/jum.16045. Epub 2022 Jun 30.
To determine the feasibility of using a deep learning (DL) algorithm to assess the quality of focused assessment with sonography in trauma (FAST) exams.
Our dataset consists of 441 FAST exams, classified as good-quality or poor-quality, with 3161 videos. We first used convolutional neural networks (CNNs), pretrained on the Imagenet dataset and fine-tuned on the FAST dataset. Second, we trained a CNN autoencoder to compress FAST images, with a 20-1 compression ratio. The compressed codes were input to a two-layer classifier network. To train the networks, each video was labeled with the quality of the exam, and the frames were labeled with the quality of the video. For inference, a video was classified as poor-quality if half the frames were classified as poor-quality by the network, and an exam was classified as poor-quality if half the videos were classified as poor-quality.
The results with the encoder-classifier networks were much better than the transfer learning results with CNNs. This was primarily because the Imagenet dataset is not a good match for the ultrasound quality assessment problem. The DL models produced video sensitivities and specificities of 99% and 98% on held-out test sets.
Using an autoencoder to compress FAST images is a very effective way to obtain features that can be used to predict exam quality. These features are more suitable than those obtained from CNNs pretrained on Imagenet.
确定使用深度学习（DL）算法评估创伤重点超声检查（FAST）检查质量的可行性。
我们的数据集包含 441 次 FAST 检查，分为高质量和低质量，共有 3161 个视频。我们首先使用在 Imagenet 数据集上预训练并在 FAST 数据集上微调的卷积神经网络（CNN）。其次，我们训练了一个 CNN 自动编码器来压缩 FAST 图像，压缩比为 20-1。压缩码被输入到一个两层分类器网络中。为了训练网络，每个视频都用检查的质量进行标记，而每一帧都用视频的质量进行标记。对于推断，如果网络将一半的帧标记为低质量，则将视频分类为低质量，如果将一半的视频分类为低质量，则将检查分类为低质量。
与 CNN 的迁移学习结果相比，编码器分类器网络的结果要好得多。这主要是因为 Imagenet 数据集与超声质量评估问题不太匹配。DL 模型在保留测试集上产生了 99%的视频灵敏度和 98%的特异性。
使用自动编码器压缩 FAST 图像是获取可用于预测检查质量的特征的一种非常有效的方法。这些特征比在 Imagenet 上预训练的 CNN 获得的特征更合适。
