Abstract
Early pulmonary nodule detection is very important in lung cancer diagnosis and screening. Most state-of-the-art lung nodule detection models are based on Faster Region-based Convolutional Neural Network (Faster R-CNN) due to its superior performance. However, this object detection approach faces difficulties with the variety of nodule sizes in training datasets. In this paper, we propose a novel Computer-Aided Detection (CAD) system based on Faster R-CNN model with adaptive anchor box for lung nodule detection. Our method employs ground-truth nodule sizes in the training dataset to generate adaptive anchor box sizes of Faster R-CNN. Learned anchors are used as hyper-parameter to boost Faster R-CNN’s detection performance. A residual convolutional neural network is proposed to reduce false positives from Faster R-CNN’s output. Our method is trained and tested on the largest publicly available LUNA16 dataset. Experiments show that our proposed system achieves a high sensitivity of 95.64% at 1.72 false positives per scan, and a Competition Performance Metric (CPM) score of 88.2%, which outperforms other recent state-of-the-art detection methods. The false positive reduction network achieves a sensitivity of 93.8%, specificity of 97.6% and accuracy of 95.7%. An additional evaluation on a completely independent SPIE-AAPM dataset demonstrates the generalization of our proposed model with 89.3% sensitivity.
Highlights
Lung cancer is one of the most popular types of cancer and accounts for almost 25% of all cancer deaths worldwide [1]
The main contributions of our work are summarized as follows: 1) In the nodule candidate detection stage, we enhance sensitivity detection performance of Faster RCNN model by employing Mean-shift clustering technique [12] to automatically learn anchor box configurations from ground-truth nodule sizes in the training dataset; 2) In the false positive reduction stage, we propose a residual convolutional neural network based on ResNet architecture [13] to improve prediction quality by reducing false positive rate with the trade-off of decreasing some sensitivity; 3) Comprehensive experiments conducted on the publicly available LUng Nodule Analysis 2016 (LUNA16) dataset [14] show the effectiveness of our proposed method, which are better than other recent state-of-the-art pulmonary nodule detectors
At a higher false positive rate, our adaptive anchor box method still achieves a 10.7% sensitivity advantage compared with the original Faster R-Convolutional Neural Networks (CNNs) 96.8% vs 86.1% at 2 False Positive (FP)/scan)
Summary
Lung cancer is one of the most popular types of cancer and accounts for almost 25% of all cancer deaths worldwide [1]. It is reported that patients can have a 63% of 5-year survival rate if diagnosed and treated at an early stage (cancer cells have not spread outside of the lung) and only 7% at a distant stage (cancer cells have spread to distant parts of the body) [2]. The subject of pulmonary nodule detection plays an important role in early diagnosis of lung cancer and in increasing the survival rate of patients. Radiologists need to inspect hundreds of CT slices for a single patient’s CT image volume. Such a manual method may burden radiologists, causing tiredness and leading to false positives (incorrectly detecting a normal tissue as nodule) or false negatives (missed detection) of pulmonary nodules [3]. Computer-aided detection (CAD) systems are proposed for the automatic detection of lung nodules in CT scan images
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
