Classification of lung nodules in CT scans using three-dimensional deep convolutional neural networks with a checkpoint ensemble method

Abstract

BackgroundAccurately detecting and examining lung nodules early is key in diagnosing lung cancers and thus one of the best ways to prevent lung cancer deaths. Radiologists spend countless hours detecting small spherical-shaped nodules in computed tomography (CT) images. In addition, even after detecting nodule candidates, a considerable amount of effort and time is required for them to determine whether they are real nodules. The aim of this paper is to introduce a high performance nodule classification method that uses three dimensional deep convolutional neural networks (DCNNs) and an ensemble method to distinguish nodules between non-nodules.MethodsIn this paper, we use a three dimensional deep convolutional neural network (3D DCNN) with shortcut connections and a 3D DCNN with dense connections for lung nodule classification. The shortcut connections and dense connections successfully alleviate the gradient vanishing problem by allowing the gradient to pass quickly and directly. Connections help deep structured networks to obtain general as well as distinctive features of lung nodules. Moreover, we increased the dimension of DCNNs from two to three to capture 3D features. Compared with shallow 3D CNNs used in previous studies, deep 3D CNNs more effectively capture the features of spherical-shaped nodules. In addition, we use an alternative ensemble method called the checkpoint ensemble method to boost performance.ResultsThe performance of our nodule classification method is compared with that of the state-of-the-art methods which were used in the LUng Nodule Analysis 2016 Challenge. Our method achieves higher competition performance metric (CPM) scores than the state-of-the-art methods using deep learning. In the experimental setup ESB-ALL, the 3D DCNN with shortcut connections and the 3D DCNN with dense connections using the checkpoint ensemble method achieved the highest CPM score of 0.910.ConclusionThe result demonstrates that our method of using a 3D DCNN with shortcut connections, a 3D DCNN with dense connections, and the checkpoint ensemble method is effective for capturing 3D features of nodules and distinguishing nodules between non-nodules.

A Two-Stage Framework for Automated Malignant Pulmonary Nodule Detection in CT Scans

“…They evaluated their proposed system on LIDC-IDRI datasets achieving a sensitivity of (94%) and specificity of(91%). Hwejin Jung et al [18] introduced a method for nodule classification, which uses three dimensional deep convolutional neural networks (3DCNNs) and checkpoint ensemble method to discriminate nodules between non-nodules, achieving a CPM score of 0.910.…”

“…CNN [24] 70.23% 4.7 74.01% 79.47% DFCNet [24] 73.14% 4.2 80.12% 81.95 RUN:Resdiual-U-Net [12] 90.9% 2 --D48 [18] 91.3% 1.6 94.8% 98.4% 3D CMixNet Faster-RCNN [17] 90% 1.1 --DeepNet [41] 84.8% 1 --ResNet [41] 86.7% 1 --ResNet+HL [41] 90.5% 1 --ESB-ALL [18] 93.3% 0.7 96.3% 99.3% Ours Deeplab-V3 plus(ex_65) + FRCNN-Inception-V2 96.4% 0.6 97% 99.4% Table 7. The confusion matrix of the Deeplab-V3 plus(ex_65) + FRCNN-Inception-V2 model.…”

Automated Lung Cancer Detection Using Artificial Intelligence (AI) Deep Convolutional Neural Networks: A Narrative Literature Review

“…In another research group Jung et al [ 12 , 15 ] model performs lower in accuracy of 96.30% in comparison to the Toğaçar et al model accuracy of 99.51%, similarly because no image augmentation and feature selection technique was used, whereas Toğaçar et al [ 13 ] used the mRMR technique with better results. Furthermore, Jung et al research team developed a three-dimensional ensemble CNN which required more training data and computational power to run [ 15 ].…”

A Novel Transfer Learning Based Approach for Pneumonia Detection in Chest X-ray Images

“…Jaiswal et al [27] used Mask-RCNN, a deep neural network, which utilizes both global and local features for pulmonary image segmentation combined with image augmentation, alongside with dropout and L2 regularization, for pneumonia identification. Jung et al [28] use a 3D deep CNN (3D DCNN) which has shortcut connections and a 3D DCNN with dense connections. The shortcuts and dense connections solve the gradient vanishing problem.…”