This post provides an overview of chest CT scan machine learning organized by clinical goal, data representation, task, and model.

这篇文章按临床目标，数据表示，任务和模型组织了胸部CT扫描机器学习的概述。

A chest CT scan is a grayscale 3-dimensional medical image that depicts the chest, including the heart and lungs. CT scans are used for the diagnosis and monitoring of many different conditions including cancer, fractures, and infections.

胸部CT扫描是描绘胸部(包括心脏和肺)的3维灰度医学图像。 CT扫描用于诊断和监视许多不同的状况，包括癌症，骨折和感染。

临床目标 (Clinical Goal)

The clinical goal refers to the medical abnormality that is the focus of the study. The following figure illustrates some example abnormalities, shown as 2D axial slices through the CT volume:

临床目标是指作为研究重点的医学异常。 下图说明了一些示例异常，显示为通过CT体积的2D轴向切片：

Many CT machine learning papers focus on lung nodules.

许多CT机器学习论文着重于肺结节 。

Other recent work has looked at pneumonia (lung infection), emphysema (a kind of lung damage that can be caused by smoking), lung cancer, or pneumothorax (air outside of the lungs rather than inside the lungs).

最近的其他工作研究了肺炎 (肺部感染)， 肺气肿 (一种可能由吸烟引起的肺损伤)， 肺癌或气胸 (肺部空气而不是肺部空气)。

I have been focused on multiple abnormality prediction, in which the model predicts 83 different abnormal findings simultaneously.

我一直致力于多个异常预测，其中该模型同时预测83个不同的异常发现 。

数据 (Data)

There are several different ways to represent CT data in a machine learning model, illustrated in this figure:

有几种不同的方法来表示机器学习模型中的CT数据，如图所示：

3D representations include a whole CT volume which is roughly 1000 x 512 x 512 pixels, and a 3D patch which can be large (e.g. half or a quarter of a whole volume) or small (e.g. 32 x 32 x 32 pixels).

3D表示包括大约1000 x 512 x 512像素的整个CT体积，以及可以大(例如，整个体积的一半或四分之一)或小(例如，32 x 32 x 32像素)的3D补丁。

2.5D representations make use of different perpendicular planes.

2.5D表示法使用不同的垂直平面。

• The axial plane is horizontal like a belt, the coronal plane is vertical like a headband or old-style headphones, and the sagittal plane is vertical like the plane of a bow and arrow in front of an archer.
  轴向平面像皮带一样水平，冠状平面像头带或老式耳机一样垂直，而矢状面像弓箭手前面的弓箭平面一样垂直。
• If we take one axial slice, one sagittal slice, and one coronal slice, and stack them up into a 3-channel image, then we have a 2.5D slice representation.
  如果我们获取一个轴向切片，一个矢状切片和一个冠状切片，并将它们堆叠成3通道图像，则我们将获得2.5D切片表示。
• If this is done with small patches, e.g. 32 x 32 pixels, then we have a 2.5D patch representation.
  如果使用小补丁(例如32 x 32像素)完成此操作，那么我们将获得2.5D补丁表示。

Finally, 2D representations are also used. This could be a full slice (e.g. 512 x 512), or a 2D patch (e.g. 16 x 16, 32 x 32, 48 x 48). These 2D slices or patches are usually from the axial view.

最后，还使用2D表示。 这可以是完整切片(例如512 x 512)或2D补丁(例如16 x 16、32 x 32、48 x 48)。 这些2D切片或面片通常是从轴向观察的。

任务 (Task)

There are many different tasks in chest CT machine learning.

胸部CT机器学习中有许多不同的任务。

The following figure illustrates a few tasks:

下图说明了一些任务：

Binary classification involves assigning a 1 or 0 to the CT representation, for the presence (1) or absence (0) of an abnormality.

二进制分类涉及为异常的存在(1)或不存在(0)给CT表示分配1或0。

Multi-class classification is for mutually exclusive categories, like different clinical subtypes of interstitial lung disease. In this case the model assigns 0 to all categories except for 1 category.

多类别分类适用于互斥类别，例如间质性肺疾病的不同临床亚型。 在这种情况下，模型会将0分配给除1个类别以外的所有类别。

Multi-label classification is for non-mutually-exclusive categories, like atelectasis (collapsed lung tissue), cardiomegaly (enlarged heart), and mass. A CT scan might have some, all, or none of these findings, and the model determines which ones if any are present.

多标签分类适用于非互斥类别，例如肺不张(肺组织塌陷)，心脏肥大(心脏扩大)和肿块。 CT扫描可能有部分，全部或没有这些发现，并且模型确定存在哪些发现。

Object detection involves predicting the coordinates of bounding boxes around abnormalities of interest.

对象检测涉及预测感兴趣异常周围的边界框的坐标。

Segmentation involves labeling every pixel, which is conceptually like “tracing the outlines of abnormalities and coloring them in.”

分割涉及标记每个像素，从概念上讲就像“追踪异常轮廓并将其着色”。

Different labels are needed to train these models. “Presence or absence” labels for abnormalities are needed to train classification models, e.g. [atelectasis=0, cardiomegaly = 1, mass = 0]. Bounding box labels are needed to train an object detection model. Segmentation masks (traced and filled in outlines) are needed to train a segmentation model. Only “presence or absence” labels are scalable to tens of thousands of CT scans, if these labels are extracted automatically from free-text radiology reports (e.g. the RAD-ChestCT data set of 36,316 CTs). Segmentation masks are the most time-consuming to obtain because they must be drawn manually on each slice; thus, segmentation studies typically use on the order of 100–1,000 CT scans.

需要不同的标签来训练这些模型。 训练分类模型需要使用“存在或不存在”的异常标签，例如[肺不张= 0，心脏肿大= 1，质量= 0]。 需要边界框标签来训练对象检测模型。 需要分割蒙版(跟踪并填充轮廓)来训练分割模型。 如果这些标签是从自由文本放射学报告中自动提取的(例如，包含36,316个CT的RAD-ChestCT数据集 )，则只有“存在或不存在”的标签才能扩展到成千上万的CT扫描。 分割蒙版是最耗时的，因为必须在每个切片上手动绘制它们。 因此，分割研究通常使用100-1,000次CT扫描。

模型 (Model)

Convolutional neural networks are the most popular machine learning model used on CT data. For a 5-minute intro to CNNs, see this article.

卷积神经网络是用于CT数据的最流行的机器学习模型。 有关CNN的5分钟介绍，请参阅本文 。

• 3D CNNs are used for whole CT volumes or 3D patches
  3D CNN用于整个CT体积或3D补丁
• 2D CNNs are used for 2.5D representations (3 channels, axial/coronal/sagittal), in the same way that 2D CNNs can take a 3-channel RGB image as input (3 channels, red/green/blue).
  2D CNN用于2.5D表示(3通道，轴向/冠状/矢状)，就像2D CNN可以将3通道RGB图像作为输入(3通道，红色/绿色/蓝色)一样。
• 2D CNNs are used for 2D slices or 2D patches.
  2D CNN用于2D切片或2D面片。

Some CNNs combine 2D and 3D convolutions. CNNs can also be “pretrained” which typically refers to first training the CNN on a natural image dataset like ImageNet and then refining the CNN’s weights on the CT data.

一些CNN结合了2D和3D卷积。 CNN也可以是“预训练”的，通常是指首先在自然图像数据集(如ImageNet)上训练CNN，然后在CT数据上细化CNN的权重。

Here is an example architecture in which a pretrained 2D CNN (ResNet18) is applied to groups of 3 adjacent slices, followed by 3D convolution:

这是一个示例架构 ，其中将预训练的2D CNN(ResNet18)应用于3个相邻切片的组，然后进行3D卷积：

间质性肺疾病分类实例 (Interstitial Lung Disease Classification Examples)

The following table includes several example studies focused on interstitial lung disease, organized by clinical goal, data, task, and model.

下表包括按临床目标，数据，任务和模型组织的，针对间质性肺疾病的几个示例研究。

• Clinical goal: these papers are all focused on interstitial lung disease. The exact classes used differ between studies. Some studies focus on clinical groupings like idiopathic pulmonary fibrosis or idiopathic non-specific interstitial pneumonia (e.g. Wang et al. 2019 and Walsh et al. 2018). Other studies focus on lung patterns like reticulation or honeycombing (e.g. Anthimopoulos et al. 2016 and Gao et al. 2016).
  临床目标：这些论文都集中于间质性肺疾病。 研究之间使用的确切类别有所不同。 一些研究侧重于临床分组，如特发性肺纤维化或特发性非特异性间质性肺炎(例如Wang等人2019和Walsh等人2018)。 其他研究集中在网状或蜂窝状等肺部模式上(例如Anthimopoulos等，2016； Gao等，2016)。
• Data: the data sets consist of 100–1,200 CTs because all of these studies rely on manual labeling of patches, slices, or pixels, which is very time-consuming. The upside of doing patch, slice, or pixel-level classification is that it provides localization information in addition to diagnostic information.
  数据：数据集包含100–1,200个CT，因为所有这些研究都依赖于手动标记斑块，切片或像素，这非常耗时。 进行补丁，切片或像素级分类的好处是，它除了提供诊断信息外，还提供定位信息。
• Task: the tasks are mostly multi-class classification, in which each patch or slice is assigned to exactly one class out of multiple possible classes.
  任务：任务主要是多类分类，其中每个补丁或切片都被分配给多个可能类中的一个类。

• Model: some of the studies use custom CNN architectures, like Wang et al. 2019 and Gao et al. 2018, whereas other studies adapt existing CNN architectures like ResNet and AlexNet.

  模型：有些研究使用了定制的CNN架构，例如Wang等。 2019和Gao等。 2018年，而其他研究调整现有CNN架构像RESNET和AlexNet 。

附加阅读 (Additional Reading)

• For a longer, more in-depth article on this topic, see Automatic Interpretation of Chest CT Scans with Machine Learning

  有关此主题的更长时间，更深入的文章，请参阅使用机器学习对胸部CT扫描进行自动解释

• For an article about machine learning in chest x-rays, which are 2D medical images of the chest rather than 3D medical images of the chest, see Automated Chest X-Ray Interpretation

  有关胸部X射线是机器的2D医学图像而不是3D胸部医学图像的机器学习文章，请参阅自动胸部X射线解释

• For more info about CNNs, see Convolutional Neural Networks in 5 minutes and How Computers See: Intro to Convolutional Neural Networks

  有关CNN的更多信息，请参阅5分钟内的卷积神经网络和《计算机的外观：卷积神经网络简介》。

• For more details about segmentation tasks, see Segmentation: U-Net, Mask R-CNN, and Medical Applications

  有关细分任务的更多详细信息，请参阅细分：U-Net，Mask R-CNN和医疗应用

• For more details about classification tasks, see Multi-label vs. Multi-class Classification: Sigmoid vs. Softmax

  有关分类任务的更多详细信息，请参阅多标签分类与多分类分类：Sigmoid与Softmax

Originally published at http://glassboxmedicine.com on August 4, 2020.

最初于 2020年8月4日 发布在 http://glassboxmedicine.com 上。