Institute of Medical Artificial Intelligence, The Second Affiliated Hospital, Xi'an Jiaotong University, Xi'an, Shaanxi, China.
J Appl Clin Med Phys. 2023 Jul;24(7):e13964. doi: 10.1002/acm2.13964. Epub 2023 Mar 16.
Automatically assessing the malignant status of lung nodules based on CTscan images can help reduce the workload of radiologists while improving their diagnostic accuracy.
Despite remarkable progress in the automatic diagnosis of pulmonary nodules by deep learning technologies, two significant problems remain outstanding. First, end-to-end deep learning solutions tend to neglect the empirical (semantic) features accumulated by radiologists and only rely on automatic features discovered by neural networks to provide the final diagnostic results, leading to questionable reliability, and interpretability. Second, inconsistent diagnosis between radiologists, a widely acknowledged phenomenon in clinical settings, is rarely examined and quantitatively explored by existing machine learning approaches. This paper solves these problems.
We propose a novel deep neural network called MS-Net, which comprises two sequential modules: A feature derivation and initial diagnosis module (FDID), followed by a diagnosis refinement module (DR). Specifically, to take advantage of accumulated empirical features and discovered automatic features, the FDID model of MS-Net first derives a range of perceptible features and provides two initial diagnoses for lung nodules; then, these results are fed to the subsequent DR module to refine the diagnoses further. In addition, to fully consider the individual and panel diagnosis opinions, we propose a new loss function called collaborative loss, which can collaboratively optimize the individual and her peers' opinions to provide a more accurate diagnosis.
We evaluate the performance of the proposed MS-Net on the Lung Image Database Consortium image collection (LIDC-IDRI). It achieves 92.4% of accuracy, 92.9% of sensitivity, and 92.0% of specificity when panel labels are the ground truth, which is superior to other state-of-the-art diagnosis models. As a byproduct, the MS-Net can automatically derive a range of semantic features of lung nodules, increasing the interpretability of the final diagnoses.
The proposed MS-Net can provide an automatic and accurate diagnosis of lung nodules, meeting the need for a reliable computer-aided diagnosis system in clinical practice.
基于 CT 扫描图像自动评估肺结节的恶性状态有助于减少放射科医生的工作量，同时提高诊断准确性。
尽管深度学习技术在肺结节自动诊断方面取得了显著进展，但仍存在两个突出问题。首先，端到端深度学习解决方案往往忽略了放射科医生积累的经验（语义）特征，仅依赖神经网络自动发现的特征提供最终诊断结果，导致可靠性和可解释性受到质疑。其次，临床环境中广泛存在的放射科医生之间诊断不一致的现象，很少被现有机器学习方法检查和定量探索。本文解决了这些问题。
我们提出了一种名为 MS-Net 的新型深度神经网络，它由两个连续模块组成：特征推导和初步诊断模块（FDID），以及诊断细化模块（DR）。具体来说，为了利用积累的经验特征和发现的自动特征，MS-Net 的 FDID 模型首先推导一系列可感知的特征，并为肺结节提供两个初步诊断；然后，将这些结果输入到后续的 DR 模块中，进一步细化诊断。此外，为了充分考虑个体和专家组的诊断意见，我们提出了一种新的损失函数，称为协作损失，可以协同优化个体和同行的意见，提供更准确的诊断。
我们在 Lung Image Database Consortium 图像集（LIDC-IDRI）上评估了所提出的 MS-Net 的性能。当专家组标签作为真实标签时，它的准确率为 92.4%，灵敏度为 92.9%，特异性为 92.0%，优于其他最先进的诊断模型。作为副产品，MS-Net 可以自动推导一系列肺结节的语义特征，增加最终诊断的可解释性。
所提出的 MS-Net 可以为肺结节提供自动和准确的诊断，满足临床实践中对可靠计算机辅助诊断系统的需求。
