Multi-scale Contextual Information Research Articles

MIFNet: A lightweight multiscale information fusion network

Semantic segmentation technique plays a crucial role in Internet of Things applications, such as industrial robotics and self-driving. Recently deep learning approaches have boosted semantic segmentation accuracy greatly. However, their comprehensive performance in terms of accuracy and efficiency is still far from satisfactory. We observe that (1) accuracy-oriented methods rely on numerous convolution layers and sophisticated architectures, which result in heavy computational complexity and usually take a long time for inference; (2) efficiency-oriented methods fail to capture the multiscale context information for discriminative representations during the feature fusion process, thus leading to suboptimal performance. Previous semantic segmentation approaches fail to address these two challenges simultaneously. To tackle the dilemma of precise segmentation and efficient inference, we propose a novel lightweight Multiscale Information Fusion Network (MIFNet). Specifically, the proposed MIFNet mainly consists of two core components, that is, Pyramid Refinement Connection Module (PRCM) and Lightweight Information Fusion Module (LIFM). The PRCM exploits skip learning to establish dependency between different stages. Meanwhile, the pyramid attention mechanism (PAM) in PRCM, which adjusts the weight of hybrid pyramid attention vector to refine spatial features of low-level, is developed to alleviate the semantic gap. Moreover, the LIFM is designed to detect objects at multiple scales from the global-local perspective. In LIFM, the proposed multiscale dense concatenation (MDC) adopts various dilated convolution to extract multiscale local context information. Extensive experimental results on benchmarks data sets demonstrate the significantly better performance of the proposed MIFNet compared with most existing state-of-the-art methods.

MFAUNet: Multiscale feature attentive U‐Net for cardiac MRI structural segmentation

The accurate and robust automatic segmentation of cardiac structures in magnetic resonance imaging (MRI) is significant in calculating cardiac clinical functional indices, and diagnosing heart diseases. Most U-Net based methods use pooling, transposed convolution, and skip connection operations to integrate the multiscale features for improved segmentation in cardiac MRI. However, this architecture lacks adequate semantic connection between the channel and spatial information, and robustness in segmenting objects with significant shape variations. In this paper, a new multiscale feature attentive U-Net for cardiac MRI structural segmentation method is proposed. An attention mechanism is adopted after concatenating the multi-level features to aggregate different scale features and determine on which features to focus. Cascade and parallel dilated convolution is also employed in the decoder blocks and skip connection is employed to enhance the ability of sensing receptive fields for multiscale context information. Furthermore, deep supervision approach with a loss function that combines the dice and cross-entropy losses to reduce overfitting and ensure better prediction is introduced. The proposed method was evaluated on three public cardiac datasets. The experimental results indicate that the method achieved competitive segmentation performance with the three datasets, which verifies the robustness and generalisability of the proposed network. In comparison with conventional U-Net methods, the model leverages attention mechanism and dilated convolution block, which increases the semantic connection between the channel and the spatial information, and improves the robustness of the right ventricle segmentation performance. From the view of the Dice scores and segmentation results, the multiscale feature attentive U-Net method is one of effective methods in segmenting cardiac MRI structures.

Contextual Information Aided Generative Adversarial Network for Low-Light Image Enhancement

Low-light image enhancement has been gradually becoming a hot research topic in recent years due to its wide usage as an important pre-processing step in computer vision tasks. Although numerous methods have achieved promising results, some of them still generate results with detail loss and local distortion. In this paper, we propose an improved generative adversarial network based on contextual information. Specifically, residual dense blocks are adopted in the generator to promote hierarchical feature interaction across multiple layers and enhance features at multiple depths in the network. Then, an attention module integrating multi-scale contextual information is introduced to refine and highlight discriminative features. A hybrid loss function containing perceptual and color component is utilized in the training phase to ensure the overall visual quality. Qualitative and quantitative experimental results on several benchmark datasets demonstrate that our model achieves relatively good results and has good generalization capacity compared to other state-of-the-art low-light enhancement algorithms.

Learning interlaced sparse Sinkhorn matching network for video super-resolution

How to effectively fuse inter- and intra-frame spatio-temporal information plays a key role in video super-resolution (VSR). Most existing works rely heavily on the accuracy of motion estimation and compensation for spatio-temporal feature alignment. However, they cannot perform well when suffering from large-scale and complex motions. To this end, this paper introduces an efficient and effective Interlaced Sparse Sinkhorn Matching (ISSM) network for VSR, which aligns supporting frames with the reference one in the feature space by learning optimal matching between image regions across frames. Specifically, the ISSM divides the input dense affinity matrix into two sparse block matrixes: one can match long-distance regions while the other can match short-distance regions, and then we leverage an efficient Sinkhorn method on each block to learn optimal matching. Moreover, we insert a residual atrous spatial pyramid pooling module before the ISSM, which can flexibly generate multi-scale features frame by frame to capture the multi-scale context information in images. The aligned features of each adjacent frame are then fed to a bidirectional temporal fusion module to capture the rich temporal information. Finally, the fused features are sent into a frame-wise dynamic reconstruction network to produce an HR frame. Extensive evaluations on three benchmark datasets demonstrate the superiority of our method over the state-of-the-art methods in terms of PSNR and SSIM.

Airport small object detection based on feature enhancement

Video object detection is essential for airport surface surveillance, but the objects on the scene are mostly small objects with low resolution, they have no obvious feature information. Due to the scale differences of the objects and the fixed receptive field on the feature maps, detectors cannot model multi-scale context information and cover all objects. In addition, although the video detection algorithm can be used as a method to solve the problem of small object detection, the temporal feature fusion method of current video detection is very dependent on the quality of a single feature map. Therefore, this paper aims to enhance the features of small objects of a single image. First, an attentional multi-scale feature fusion enhancement (A-MSFFE) network is built on the memory-enhanced global-local aggregation (MEGA) to supplement semantic and spatial information of small objects. Then, a context feature enhancement (CFE) module is designed for obtaining different receptive fields through different dilated convolutions. Meanwhile, a video detection dataset about the airport is established. Finally, the experimental results show that the proposed method can improve the detection accuracies of small objects and outperform other state-of-the-art video object detection algorithms in self-built airport dataset.

Parallel multiscale context-based edge-preserving optical flow estimation with occlusion detection

Although convolutional neural network (CNN)-based optical flow approaches have exhibited good performance in terms of computational accuracy and efficiency in recent years, the issue of edge-blurring caused by motion occlusions remains. In this paper, we propose a parallel multiscale context-based edge-preserving optical flow estimation method with occlusion detection, named PMC-PWC. First, we exploit a parallel multiscale context (PMC) network for occlusion detection, in which the proposed PMC model is able to aggregate the multiscale context information to develop the performance of occlusion detection near motion boundaries. Second, we combine the PMC model with a context network to plan an occlusion estimation module and incorporate it into a pyramid, warping, and cost volume model to construct an edge-preserving optical flow computation network. Third, we design a novel loss function including an endpoint error (EPE)-based loss, a binary cross-entropy loss and an edge loss to supervise the proposed PMC-PWC network to produce optical flow and occlusion simultaneously. Finally, we run the proposed PMC-PWC method on the MPI-Sintel and KITTI datasets to conduct a comprehensive comparison with several state-of-the-art approaches. The experimental results indicate that the proposed PMC-PWC method performed well in terms of both accuracy and robustness, especially due to the significant benefits of edge preservation and occlusion handling.

EMS-Net: Enhanced Multi-Scale Network for Polyp Segmentation.

In recent years, polyp segmentation plays an important role in the diagnosis and treatment of colorectal cancer. Accurate segmentation of polyps is very challenging due to different sizes, shapes, and unclear boundaries. Making full use of multi-scale contextual information to segment polyps may bring better results. In this paper, we propose an enhanced multi-scale network for accurate polyp segmentation. It is composed of a multi-scale connected baseline (U-Net+++), a multi-scale backbone (Res2Net), three Receptive Field Block (RFB) modules, and four Local Context Attention (LCA) modules. Specifically, the baseline's multi-scale skip connections can aggregate features in both low-level and high-level layers. We have evaluated our model on three publicly available and challenging datasets (EndoScene, CVC-ClinicDB, Kvasir-SEG). Compared with other methods, our model achieves SOTA performance. It is noteworthy that our model is the only network that has achieved over 0.900 mean Dice on EndoScene and CVC-ClinicDB.

Liver tumor segmentation using a new asymmetrical dilated convolutional semantic segmentation network in CT images

AbstractLiver cancer has become one of the commonly occurring cancers in men and women. The segmented computed tomography (CT) image of the tumor is the major source for a doctor to diagnose liver diseases. This article proposes a fully automatic system based on an asymmetric dilated convolutional encoder–decoder neural (ADCEDN) network for the problem of accurately extracting the liver and tumor from abdominal CT slices. Specifically, two ADCEDN networks are used in a cascaded form to perform both liver and liver abnormality segmentation. We enrich the encoder module by incorporating the multiscale contextual information through dilated convolutions that enhance the capacity and efficiency of the network. Hybrid dilated convolution across depths perform depth‐wise separable convolution by providing a less steep increase in dilation rate and thus denser sampling that can reduce the computation cost and several metrics while maintaining better performance. Also, the encoder–decoder network is made asymmetric, thereby making the network less complex and more generalizable. The suggested method produced an average dice similarity coefficient score of 96.54% and 76.16% for liver segmentation and liver abnormality segmentation on the 3D Image Reconstruction for Comparison of Algorithm Database data set and 96.42% and 75.7% on the Liver Tumor Segmentation data set. The test outcomes show that the proposed ADCEDN attains state‐of‐the‐art performance in both segmentation of liver and tumors.

Multi-Supervised Encoder-Decoder for Image Forgery Localization

Image manipulation localization is one of the most challenging tasks because it pays more attention to tampering artifacts than to image content, which suggests that richer features need to be learned. Unlike many existing solutions, we employ a semantic segmentation network, named Multi-Supervised Encoder–Decoder (MSED), for the detection and localization of forgery images with arbitrary sizes and multiple types of manipulations without extra pre-training. In the basic encoder–decoder framework, the former encodes multi-scale contextual information by atrous convolution at multiple rates, while the latter captures sharper object boundaries by applying upsampling to gradually recover the spatial information. The additional multi-supervised module is designed to guide the training process by multiply adopting pixel-wise Binary Cross-Entropy (BCE) loss after the encoder and each upsampling. Experiments on four standard image manipulation datasets demonstrate that our MSED network achieves state-of-the-art performance compared to alternative baselines.

Evaluation of Deep Learning Segmentation Models for Detection of Pine Wilt Disease in Unmanned Aerial Vehicle Images

Pine wilt disease (PWD) is a serious threat to pine forests. Combining unmanned aerial vehicle (UAV) images and deep learning (DL) techniques to identify infected pines is the most efficient method to determine the potential spread of PWD over a large area. In particular, image segmentation using DL obtains the detailed shape and size of infected pines to assess the disease’s degree of damage. However, the performance of such segmentation models has not been thoroughly studied. We used a fixed-wing UAV to collect images from a pine forest in Laoshan, Qingdao, China, and conducted a ground survey to collect samples of infected pines and construct prior knowledge to interpret the images. Then, training and test sets were annotated on selected images, and we obtained 2352 samples of infected pines annotated over different backgrounds. Finally, high-performance DL models (e.g., fully convolutional networks for semantic segmentation, DeepLabv3+, and PSPNet) were trained and evaluated. The results demonstrated that focal loss provided a higher accuracy and a finer boundary than Dice loss, with the average intersection over union (IoU) for all models increasing from 0.656 to 0.701. From the evaluated models, DeepLLabv3+ achieved the highest IoU and an F1 score of 0.720 and 0.832, respectively. Also, an atrous spatial pyramid pooling module encoded multiscale context information, and the encoder–decoder architecture recovered location/spatial information, being the best architecture for segmenting trees infected by the PWD. Furthermore, segmentation accuracy did not improve as the depth of the backbone network increased, and neither ResNet34 nor ResNet50 was the appropriate backbone for most segmentation models.

Multi-scale context extractor network for water-body extraction from high-resolution optical remotely sensed images

Water-body surveying and mapping is of great significance for water resources utilization, flood monitoring, and environmental protection. However, due to distribution diversities, shape and size variations, and complex scenarios of water-bodies, it is still challengeable to accurately and efficiently extract water-bodies from high-resolution remotely sensed images. In this paper, we propose a multi-scale context extractor network, termed as MSCENet, for delineating water-bodies from high-resolution optical remotely sensed images. The MSCENet mainly contains three key parts: a multi-scale feature encoder, a feature decoder, and a context feature extractor module. To address shape and size variations of water-bodies, the Res2Net is used in the feature encoder to extract rich multi-scale information of water-bodies. The context extractor module is composed of an assorted dilated convolution unit and a complex multi-kernel pooling unit, which further extracts multi-scale contextual information to generate high-level feature maps. The robustness and effectiveness of our MSCENet have been evaluated on two public datasets: LandCover.ai Data Set and DeepGlobe Data Set. Comparative experiments indicate the superiority and applicability of the MSCENet in water-body extraction.

FD-SSD: An improved SSD object detection algorithm based on feature fusion and dilated convolution

Objects that occupy a small portion of an image or a frame contain fewer pixels and contains less information. This makes small object detection a challenging task in computer vision. In this paper, an improved Single Shot multi-box Detector based on feature fusion and dilated convolution (FD-SSD) is proposed to solve the problem that small objects are difficult to detect. The proposed network uses VGG-16 as the backbone network, which mainly includes a multi-layer feature fusion module and a multi-branch residual dilated convolution module. In the multi-layer feature fusion module, the last two layers of the feature map are up-sampled, and then they are concatenated at the channel level with the shallow feature map to enhance the semantic information of the shallow feature map. In the multi-branch residual dilated convolution module, three dilated convolutions with different dilated ratios based on the residual network are combined to obtain the multi-scale context information of the feature without losing the original resolution of the feature map. In addition, deformable convolution is added to each detection layer to better adapt to the shape of small objects. The proposed FD-SSD achieved 79.1% mAP and 29.7% mAP on PASCAL VOC2007 dataset and MS COCO dataset respectively. Experimental results show that FD-SSD can effectively improve the utilization of multi-scale information of small objects, thus significantly improve the effect of the small object detection.

GCAUNet: A group cross-channel attention residual UNet for slice based brain tumor segmentation

Precise brain tumor segmentation can improve patient prognosis. However, due to the complicated structure of the human brain, brain tumor segmentation is a challenging task. To improve the brain tumor segmentation performance, a group cross-channel attention residual UNet (GCAUNet) that can make full use of the low-level fine details of tumor regions is proposed. First, iterative background removal and image normalization are applied to remove the disturbances of the background and illumination variations, respectively. Then, GCAUNet is constructed for brain tumor segmentation. To recover the fine details of brain tumors, a parallel network path, namely, the detail recovering (DR) path, is introduced to extract detail feature groups from multiscale low-level feature maps. In addition, to emphasize the significant feature groups and channels, a coarse-to-fine cross channel attention module, namely, the group cross-channel attention (GCA) module, is proposed. Furthermore, a multiscale input (MI) path is introduced into GCAUNet to acquire multiscale context information. The experimental results show that the average dices of GCAUNet for the whole tumor (WT), tumor core (TC) and enhancing tumor (ET) on BraTS 2017 and BraTS 2018 reach 87.2%/79.6%/78.1% and 90.9%/84.5%/81.3%, respectively. Compared with the backbone, the dices of WT, TC and ET on BraTS 2017 and 2018 are improved by 4.8%/7.1%/6.2% and 4.1%/3.5%/3.1%, respectively, which indicates that GCAUNet can significantly improve the brain tumor segmentation performance.

Fully automated segmentation of brain tumor from multiparametric MRI using 3D context deep supervised U-Net.

Owing to histologic complexities of brain tumors, its diagnosis requires the use of multimodalities to obtain valuable structural information so that brain tumor subregions can be properly delineated. In current clinical workflow, physicians typically perform slice-by-slice delineation of brain tumor subregions, which is a time-consuming process and also more susceptible to intra- and inter-rater variabilities possibly leading to misclassification. To deal with this issue, this study aims to develop an automatic segmentation of brain tumor in MR images using deep learning. In this study, we develop a context deep-supervised U-Net to segment brain tumor subregions. A context block which aggregates multiscale contextual information for dense segmentation was proposed. This approach enlarges the effective receptive field of convolutional neural networks, which, in turn, improves the segmentation accuracy of brain tumor subregions. We performed the fivefold cross-validation on the Brain Tumor Segmentation Challenge (BraTS) 2020 training dataset. The BraTS 2020 testing datasets were obtained via BraTS online website as a hold-out test. For BraTS, the evaluation system divides the tumor into three regions: whole tumor (WT), tumor core (TC), and enhancing tumor (ET). The performance of our proposed method was compared against two state-of-the-arts CNN networks in terms of segmentation accuracy via Dice similarity coefficient (DSC) and Hausdorff distance (HD). The tumor volumes generated by our proposed method were compared with manually contoured volumes via Bland-Altman plots and Pearson analysis. The proposed method achieved the segmentation results with a DSC of 0.923±0.047, 0.893±0.176, and 0.846±0.165 and a 95% HD95 of 3.946±7.041, 3.981±6.670, and 10.128±51.136mm on WT, TC, and ET, respectively. Experimental results demonstrate that our method achieved comparable to significantly (p<0.05) better segmentation accuracies than other two state-of-the-arts CNN networks. Pearson correlation analysis showed a high positive correlation between the tumor volumes generated by proposed method and manual contour. Overall qualitative and quantitative results of this work demonstrate the potential of translating proposed technique into clinical practice for segmenting brain tumor subregions, and further facilitate brain tumor radiotherapy workflow.

CDADNet: Context-guided dense attentional dilated network for crowd counting

Crowd counting is a conspicuous task in computer vision owing to scale variations, perspective distortions, and complex backgrounds. Existing research usually adopts the dilated convolution network to enlarge the receptive fields to solve the problem of scale variations. However, these methods easily bring background information into the large receptive fields to generate poor quality density maps. To address this problem, we propose a novel backbone called Context-guided Dense Attentional Dilated Network (CDADNet). CDADNet contains three components: an attentional module, a context-guided module and a dense attentional dilated module. The attentional module is used to provide attention maps which can remove background information, while the context-guided module is proposed to extract multi-scale contextual information. Moreover, the dense attentional dilated module aims to generate high-granularity density maps and the cascaded strategy is used to preserve information from changing scales. To verify the feasibility of our method, we compare it to the existing approaches on five crowd counting datasets (ShanghaiTech (Part_A and Part_B), WorldEXPO’10, UCSD, UCF_CC_50). The comparison results demonstrate that CDADNet is effective and robust for various scenes.

Edge and neighborhood guidance network for 2D medical image segmentation

Accurate automatic image segmentation is important in medical image analysis. A perfect segmentation using fully convolutional network (FCN) means an accurate classification of each pixel. However, it is still a great challenge to accurately differentiate edge pixels from neighborhood pixels in weak edge regions. Many previous segmentation methods have focused on edge information to mitigate weak edge problems, but the more important neighborhood information is undervalued. To tackle this problem, in this paper, we propose a novel yet effective Edge and Neighborhood Guidance Network (ENGNet). Specifically, instead of just utilizing the edge information as the shape constraints, the edge and neighborhood guidance (ENG) module is designed to exploit the edge information and fine-grained neighborhood spatial information simultaneously, so as to improve the ability of network to classify edge pixels and neighborhood pixels in weak edge regions. Moreover, the ENG modules are adopted in different scales to learn sufficient feature representations of edge and neighborhood. To extract complementary features more effectively in channel dimension, we also design a multi-scale adaptive selection (MAS) module at channel-wise to extract multi-scale context information and adaptively fuse different-scale features. Two 2D public segmentation datasets including skin lesion dataset and endoscopic polyp dataset are used to evaluate the performance of the proposed ENGNet. Experimental results demonstrated that by exploiting edge information and neighborhood spatial information in different scales simultaneously, the proposed ENGNet can effectively alleviate the misclassification in weak edge regions and achieve better performance than other state-of-the-art methods.

Improved 3D U-Net for COVID-19 Chest CT Image Segmentation

Coronavirus disease 2019 (COVID-19) has spread rapidly worldwide. The rapid and accurate automatic segmentation of COVID-19 infected areas using chest computed tomography (CT) scans is critical for assessing disease progression. However, infected areas have irregular sizes and shapes. Furthermore, there are large differences between image features. We propose a convolutional neural network, named 3D CU-Net, to automatically identify COVID-19 infected areas from 3D chest CT images by extracting rich features and fusing multiscale global information. 3D CU-Net is based on the architecture of 3D U-Net. We propose an attention mechanism for 3D CU-Net to achieve local cross-channel information interaction in an encoder to enhance different levels of the feature representation. At the end of the encoder, we design a pyramid fusion module with expanded convolutions to fuse multiscale context information from high-level features. The Tversky loss is used to resolve the problems of the irregular size and uneven distribution of lesions. Experimental results show that 3D CU-Net achieves excellent segmentation performance, with Dice similarity coefficients of 96.3% and 77.8% in the lung and COVID-19 infected areas, respectively. 3D CU-Net has high potential to be used for diagnosing COVID-19.

RetFluidNet: Retinal Fluid Segmentation for SD-OCT Images Using Convolutional Neural Network.

Age-related macular degeneration (AMD) is one of the leading causes of irreversible blindness and is characterized by fluid-related accumulations such as intra-retinal fluid (IRF), subretinal fluid (SRF), and pigment epithelial detachment (PED). Spectral-domain optical coherence tomography (SD-OCT) is the primary modality used to diagnose AMD, yet it does not have algorithms that directly detect and quantify the fluid. This work presents an improved convolutional neural network (CNN)-based architecture called RetFluidNet to segment three types of fluid abnormalities from SD-OCT images. The model assimilates different skip-connect operations and atrous spatial pyramid pooling (ASPP) to integrate multi-scale contextual information; thus, achieving the best performance. This work also investigates between consequential and comparatively inconsequential hyperparameters and skip-connect techniques for fluid segmentation from the SD-OCT image to indicate the starting choice for future related researches. RetFluidNet was trained and tested on SD-OCT images from 124 patients and achieved an accuracy of 80.05%, 92.74%, and 95.53% for IRF, PED, and SRF, respectively. RetFluidNet showed significant improvement over competitive works to be clinically applicable in reasonable accuracy and time efficiency. RetFluidNet is a fully automated method that can support early detection and follow-up of AMD.

FE-RetinaNet: Small Target Detection with Parallel Multi-Scale Feature Enhancement

Because small targets have fewer pixels and carry fewer features, most target detection algorithms cannot effectively use the edge information and semantic information of small targets in the feature map, resulting in low detection accuracy, missed detections, and false detections from time to time. To solve the shortcoming of insufficient information features of small targets in the RetinaNet, this work introduces a parallel-assisted multi-scale feature enhancement module MFEM (Multi-scale Feature Enhancement Model), which uses dilated convolution with different expansion rates to avoid multiple down sampling. MFEM avoids information loss caused by multiple down sampling, and at the same time helps to assist shallow extraction of multi-scale context information. Additionally, this work adopts a backbone network improvement plan specifically designed for target detection tasks, which can effectively save small target information in high-level feature maps. The traditional top-down pyramid structure focuses on transferring high-level semantics from the top to the bottom, and the one-way information flow is not conducive to the detection of small targets. In this work, the auxiliary MFEM branch is combined with RetinaNet to construct a model with a bidirectional feature pyramid network, which can effectively integrate the strong semantic information of the high-level network and high-resolution information regarding the low level. The bidirectional feature pyramid network designed in this work is a symmetrical structure, including a top-down branch and a bottom-up branch, performs the transfer and fusion of strong semantic information and strong resolution information. To prove the effectiveness of the algorithm FE-RetinaNet (Feature Enhancement RetinaNet), this work conducts experiments on the MS COCO. Compared with the original RetinaNet, the improved RetinaNet has achieved a 1.8% improvement in the detection accuracy (mAP) on the MS COCO, and the COCO AP is 36.2%; FE-RetinaNet has a good detection effect on small targets, with APs increased by 3.2%.

Automatic lung nodule detection in thoracic CT scans using dilated slice-wise convolutions.

Most state-of-the-art automated medical image analysis methods for volumetric data rely on adaptations of two-dimensional (2D) and three-dimensional (3D) convolutional neural networks (CNNs). In this paper, we develop a novel unified CNN-based model that combines the benefits of 2D and 3D networks for analyzing volumetric medical images. In our proposed framework, multiscale contextual information is first extracted from 2D slices inside a volume of interest (VOI). This is followed by dilated 1D convolutions across slices to aggregate in-plane features in a slice-wise manner and encode the information in the entire volume. Moreover, we formalize a curriculum learning strategy for a two-stage system (i.e., a system that consists of screening and false positive reduction), where the training samples are presented to the network in a meaningful order to further improve the performance. We evaluated the proposed approach by developing a computer-aided detection (CADe) system for lung nodules. Our results on 888 CT exams demonstrate that the proposed approach can effectively analyze volumetric data by achieving a sensitivity of > 0.99 in the screening stage and a sensitivity of > 0.96 at eight false positives per case in the false positive reduction stage. Our experimental results show that the proposed method provides competitive results compared to state-of-the-art 3D frameworks. In addition, we illustrate the benefits of curriculum learning strategies in two-stage systems that are of common use in medical imaging applications.