Lu Yaozhi, Aslani Shahab, Emberton Mark, Alexander Daniel C, Jacob Joseph
Centre for Medical Image Computing, University College London, London WC1V 6LJ, U.K.
Department of Computer Science, University College London, London WC1E 6BT, U.K.
IEEE Access. 2022;10:34369-34378. doi: 10.1109/ACCESS.2022.3161954.
In this study, the long-term mortality in the National Lung Screening Trial (NLST) was investigated using a deep learning-based method. Binary classification of the non-lung-cancer mortality (i.e. cardiovascular and respiratory mortality) was performed using neural network models centered around a 3D-ResNet. The models were trained on a participant age, gender, and smoking history matched cohort. Utilising both the 3D CT scan and clinical information, the models can achieve an AUC of 0.73 which outperforms humans at cardiovascular mortality prediction. The corresponding F1 and Matthews Correlation Coefficient are 0.60 and 0.38 respectively. By interpreting the trained models with 3D saliency maps, we examined the features on the CT scans that correspond to the mortality signal. By extracting information from 3D CT volumes, we can highlight regions in the thorax region that contribute to mortality that might be overlooked by the clinicians. Therefore, this can help focus preventative interventions appropriately, particularly for under-recognised pathologies and thereby reducing patient morbidity.
在本研究中，使用基于深度学习的方法对国家肺癌筛查试验（NLST）中的长期死亡率进行了调查。使用以3D-ResNet为核心的神经网络模型对非肺癌死亡率（即心血管和呼吸系统死亡率）进行二元分类。这些模型在参与者年龄、性别和吸烟史匹配的队列上进行训练。利用3D CT扫描和临床信息，这些模型可以实现0.73的AUC，在心血管死亡率预测方面优于人类。相应的F1和马修斯相关系数分别为0.60和0.38。通过用3D显著性图解释训练好的模型，我们检查了CT扫描上与死亡率信号相对应的特征。通过从3D CT体积中提取信息，我们可以突出胸部区域中对死亡率有贡献的区域，而这些区域可能会被临床医生忽视。因此，这有助于适当地集中预防性干预措施，特别是针对未被充分认识的病症，从而降低患者的发病率。
