Volumetric lung nodule segmentation using adaptive ROI with multi-view residual learning

Abstract

Byung-il Lee 2 & Yeong-Gil Shin 1 Accurate quantification of pulmonary nodules can greatly assist the early diagnosis of lung cancer, enhancing patient survival possibilities. A number of nodule segmentation techniques, which either rely on a radiologist-provided 3-D volume of interest (VOI) or use the constant region of interests (ROIs) for all the slices, are proposed; however, these techniques can only investigate the presence of nodule voxels within the given VOI. Such approaches restrain the solutions to freely investigate the nodule presence outside the given VOI and also include the redundant structures (non-nodule) into VOI, which limits the segmentation accuracy. In this work, a novel semi-automated approach for 3-D segmentation of lung nodule in computerized tomography scans, has been proposed. The technique is segregated into two stages. In the first stage, a 2-D ROI containing the nodule is provided as an input to perform a patch-wise exploration along the axial axis using a novel adaptive ROI algorithm. This strategy enables the dynamic selection of the ROI in the surrounding slices to investigate the presence of nodules using a Deep Residual U-Net architecture. This stage provides the initial estimation of the nodule utilized to extract the VOI. In the second stage, the extracted VOI is further explored along the coronal and sagittal axes, in patchwise fashion, with Residual U-Nets. All the estimated masks are then fed into a consensus module to produce a final volumetric segmentation of the nodule. The algorithm is rigorously evaluated on LIDC-IDRI dataset, which is the largest publicly available dataset. The proposed approach achieved the average dice score of 87.5%, which is significantly higher than the existing state-of-the-art techniques. Lung cancer is one of the most severe and highly-prevalent cancers and is the leading cause of cancer deaths worldwide 1. It has been forecasted to be one of the greatest single cause of mortality among the European population in 2019 2. Early diagnosis of lung cancer is crucial to enable possible life-saving interventions 3 , which relies on accurate quantification of pulmonary nodule; albeit pulmonary nodules can be associated with several diseases, their recurrent diagnosis is lung cancer. The continuous monitoring of lung nodule volume is vital to estimate the malignancy and to better forecast, the probability of lung cancer 45. For calculation of volume, the nodule is first segmented, while the manual segmentation of nodule is a tedious and time-consuming task which also introduces the inter and intra-observer variabilities 6. Computer-aided diagnosis (CAD) systems have huge potential to overcome the challenges faced during manual segmentation of pulmonary nodules and can remarkably enhance the productivity of radiologists. Therefore, several automatic nodule segmentation techniques have been proposed to facilitate radiologists, including advanced deep learning and classical image processing based techniques 7. All existing techniques require a 3-D volum...

Deep Learning Applications in Computed Tomography Images for Pulmonary Nodule Detection and Diagnosis: A Review

“…U-Net and Fully Convolutional Neural Networks (FCN) architectures are two basic structures that are frequently used. Numerous works have shown that convolutional neural networks architecture can significantly improve the performance of lung segmentation [ 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 ], especially semantic segmentation networks such as FCN [ 61 ] and U-Net [ 62 ]. Such networks implement two key steps.…”

“…The proposed model was trained and validated on the LUNA16 dataset and achieved 0.736 DSC. Usman et al [ 56 ] proposed a dynamic modification region of interest (ROI) algorithm. This approach used Deep Res-UNet as the foundation for locating the input lung nodule volumes and improving lung nodule segmentation.…”

Leveraging Data Science to Combat COVID-19: A Comprehensive Review

“…Various studies [120]- [122] have used computer vision algorithms to speed up the process of disease detection across several imaging modalities with some studies demonstrating that image analysis techniques have the potential to outperform expert radiologists [123], [124]. To diagnose COVID-19, two medical imaging modalities (CT and X-ray) have been experimented with [125], which we discuss below.…”

Optical Flow Methods for Lung Nodule Segmentation on LIDC-IDRI Images

“…Detection of a lung nodule, segmentation and classification only based on simple morphological and textural properties such as size or texture or shape features is not robust as suggested by Paing et al [38] and does not reveal the exact magnitude of the underlying challenges in lung cancer detection and diagnosis. Researchers have used various techniques for lung nodule detection and segmentation such as 3D tensor filtering with local image feature analysis [18], global optimal active contour model [54], corner seeded region growing combined with differential evolution based optimal thresholding [35], connected component labelling with morphological operations and multilayer perceptron [21], sparse field level sets and boosting algorithm [39], adaptive ROI with multi-view residual learning [49], LBF active contour model with information entropy and joint vector [22] etc. including optical flow methods for evaluation of interval change in metastatic lung nodules [17], and segmentation by background subtraction [45].…”