基于胸部X射线影像的常见疾病预测方法研究

Prediction Method for Common Diseases Based on Chest X-Ray Images

利用X射线影像进行胸部疾病诊断是一种常用的诊断方法,具有重要的临床诊断价值。随着大规模可用数据集的发布,已经提出了几种利用胸部X射线图像预测常见疾病的方法。然而大多数现有的预测模型大都仅考虑单个视图,忽略了多视图影像对于临床医生诊断的支持作用。此外使用单个模型进行影像特征抽取时,存在有效特征提取不全的问题,进而导致疾病预测准确率较低。为此,提出了一种新的深度相关多级特征融合方法(DFFM),该方法融合不同模型提取的不同视图的视觉特征,以提高疾病预测的准确性。并在目前最大的胸部X射线数据集MIMIC-CXR上进行了验证,实验结果表明,所提方法的area under the receiver operating characteristic curve(AUC)值达0.847,与现有的单视图及简单进行特征拼接的多视图模型相比,AUC值分别提升了12.6个百分点和5.3个百分点,验证了所提多级融合方法的有效性。

Xray imaging is a commonly used diagnostic method with important clinical value in chestdisease diagnosis.Exploiting the release of largescale available datasets,several methods have been proposed for predicting common diseases using chest Xray images.However,most of the existing predictive models are limited to singleview inputs,ignoring the supportive role of multiview images in clinical diagnosis.Additionally when image features are extracted using a single model,the effective features are incompletely extracted and the accuracy of disease prediction decreases.The present study proposes a new depthdependent multilevel feature fusion method(DFFM)that combines the visual features of different views extracted via different models to improve the accuracy of disease prediction.DFFM was verified using MIMICCXR,the largest available chest Xray dataset.Experimental results show that the area under the receiver operating characteristic curve was 0.847,12.6 and 5.3 percentage points higher than the existing singleview and multiview models with simple feature splicing,respectively.These results confirm the effectiveness of the proposed multilevel fusion method.