Ferreira Jose Raniery, Armando Cardona Cardenas Diego, Moreno Ramon Alfredo, de Fatima de Sa Rebelo Marina, Krieger Jose Eduardo, Antonio Gutierrez Marco
Annu Int Conf IEEE Eng Med Biol Soc. 2020 Jul;2020:1238-1241. doi: 10.1109/EMBC44109.2020.9176517.
Pneumonia is one of the leading causes of childhood mortality worldwide. Chest x-ray (CXR) can aid the diagnosis of pneumonia, but in the case of low contrast images, it is important to include computational tools to aid specialists. Deep learning is an alternative because it can identify patterns automatically, even in low-resolution images. We propose herein a convolutional neural network (CNN) architecture with different training strategies towards detecting pneumonia on CXRs and distinguishing its subforms of bacteria and virus. We also evaluated different image pre-processing methods to improve the classification. This study used CXRs from pediatric patients from a public pneumonia CXR dataset. The pre-processing methods evaluated were image cropping and histogram equalization. To classify the images, we adopted the VGG16 CNN and replaced its fully-connected layers with a customized multilayer perceptron. With this architecture, we proposed and evaluated four different training strategies: original CXR image (baseline), chest-cavity-cropped image (A), and histogram-equalized segmented image (B). The last strategy method (C) implemented is based on ensemble between strategies A and B. The performance was assessed by the area under the ROC curve (AUC) with 95% confidence interval (CI), accuracy, sensitivity, specificity, and F1-score. The ensemble model C yielded the highest performances: AUC of 0.97 (CI: 0.96-0.99) to classify pneumonia vs. normal, and AUC of 0.91 (CI: 0.88-0.94) to classify bacterial vs. viral cases. All models that used pre-processed images showed higher AUC than baseline, which used the original CXR image. Image cropping and histogram equalization reduced irrelevant information from the exam, enhanced contrast, and was able to identify fine CXR texture details. The proposed ensemble model increased the representation of inflammatory patterns from bacteria and viruses with few epochs to train the deep CNNs.Clinical relevance- Deep learning can identify complex radiographic patterns in low contrast images due to pneumonia and distinguish its subforms of bacteria and virus. The correlation of imaging with lab results could accelerate the adoption of complementary exams to confirm the disease's cause.
肺炎是全球儿童死亡的主要原因之一。胸部X光（CXR）有助于肺炎的诊断，但在图像对比度较低的情况下，纳入计算工具以协助专家诊断很重要。深度学习是一种选择，因为它能够自动识别模式，即使是在低分辨率图像中。我们在此提出一种卷积神经网络（CNN）架构，采用不同的训练策略来检测CXR上的肺炎，并区分其细菌和病毒亚型。我们还评估了不同的图像预处理方法以改进分类。本研究使用了来自一个公共肺炎CXR数据集的儿科患者的CXR图像。评估的预处理方法包括图像裁剪和直方图均衡化。为了对图像进行分类，我们采用了VGG16 CNN，并将其全连接层替换为定制的多层感知器。基于这种架构，我们提出并评估了四种不同的训练策略：原始CXR图像（基线）、胸腔裁剪图像（A）和直方图均衡化分割图像（B）。最后实施的策略方法（C）是基于策略A和B的集成。通过ROC曲线下面积（AUC）以及95%置信区间（CI）、准确率、灵敏度、特异性和F1分数来评估性能。集成模型C表现出最高的性能：分类肺炎与正常情况时AUC为0.97（CI：0.96 - 0.99），分类细菌与病毒病例时AUC为0.91（CI：0.88 - 0.94）。所有使用预处理图像的模型的AUC都高于使用原始CXR图像的基线模型。图像裁剪和直方图均衡化减少了检查中的无关信息，增强了对比度，并能够识别CXR的精细纹理细节。所提出的集成模型在训练深度CNN时只需较少轮次就能增加细菌和病毒炎症模式的表征。临床相关性——深度学习能够识别因肺炎导致的低对比度图像中的复杂放射学模式，并区分其细菌和病毒亚型。影像学与实验室结果的相关性可以加速采用辅助检查来确定疾病病因。
