Task For Computer-aided Diagnosis Research Articles

Boundary-Aware Gradient Operator Network for Medical Image Segmentation.

Medical image segmentation is a crucial task in computer-aided diagnosis. Although convolutional neural networks (CNNs) have made significant progress in the field of medical image segmentation, the convolution kernels of CNNs are optimized from random initialization without explicitly encoding gradient information, leading to a lack of specificity for certain features, such as blurred boundary features. Furthermore, the frequently applied down-sampling operation also loses the fine structural features in shallow layers. Therefore, we propose a boundary-aware gradient operator network (BG-Net) for medical image segmentation, in which the gradient convolution (GConv) and the boundary-aware mechanism (BAM) modules are developed to simulate image boundary features and the remote dependencies between channels. The GConv module transforms the gradient operator into a convolutional operation that can extract gradient features; it attempts to extract more features such as images boundaries and textures, thereby fully utilizing limited input to capture more features representing boundaries. In addition, the BAM can increase the amount of global contextual information while suppressing invalid information by focusing on feature dependencies and the weight ratios between channels. Thus, the boundary perception ability of BG-Net is improved. Finally, we use a multi-modal fusion mechanism to effectively fuse lightweight gradient convolution and U-shaped branch features into a multilevel feature, enabling global dependencies and low-level spatial details to be effectively captured in a shallower manner. We conduct extensive experiments on eight datasets that broadly cover medical images to evaluate the effectiveness of the proposed BG-Net. The experimental results demonstrate that BG-Net outperforms the state-of-the-art methods, particularly those focused on boundary segmentation. The codes are available at https://github.com/LiYu51/BG-Net.

SAH-NET: Structure-Aware Hierarchical Network for Clustered Microcalcification Classification in Digital Breast Tomosynthesis.

Benign and malignant classification of clustered microcalcifications (MCs) in digital breast tomosynthesis (DBT) is an essential task in computer-aided diagnosis. However, due to the anisotropic resolution of DBT, three-dimensional (3-D) convolutional neural network (CNN)-based methods cannot extract hierarchical features efficiently. Moreover, the sparse distribution of MC points in the cluster makes it difficult for the CNN to extract discriminative structural information for classification. To comprehensively address these challenges, we propose a novel structure-aware hierarchical network (SAH-Net) for benign and malignant classification of clustered MC in a DBT volume. Specifically, the two-dimensional (2-D) group convolution is used to extract intraslice features. The one-to-one correspondence between group convolutions and slices ensures the independence of hierarchical feature extraction. Then, a partial deformable Transformer-based 3-D structural feature learning module is proposed to capture the long-range dependency between MC points in the cluster. We evaluate the proposed method on an in-house dataset with 495 clustered MCs collected from 462 DBT images. Experimental results confirm the validity of our proposed modules. The results also show that the proposed SAH-Net outperforms several other representative methods on this topic, and achieves the best classification result, with an area under the receiver operation curve (AUC) of 86.87%. The implementation of the proposed model is available at https://github.com/sunhaotian130911/SAHNet.

LGDNet: local feature coupling global representations network for pulmonary nodules detection.

Detection of suspicious pulmonary nodules from lung CT scans is a crucial task in computer-aided diagnosis (CAD) systems. In recent years, various deep learning-based approaches have been proposed and demonstrated significant potential for addressing this task. However, existing deep convolutional neural networks exhibit limited long-range dependency capabilities and neglect crucial contextual information, resulting in reduced performance on detecting small-size nodules in CT scans. In this work, we propose a novel end-to-end framework called LGDNet for the detection of suspicious pulmonary nodules in lung CT scans by fusing local features and global representations. To overcome the limited long-range dependency capabilities inherent in convolutional operations, a dual-branch module is designed to integrate the convolutional neural network (CNN) branch that extracts local features with the transformer branch that captures global representations. To further address the issue of misalignment between local features and global representations, an attention gate module is proposed in the up-sampling stage to selectively combine misaligned semantic data from both branches, resulting in more accurate detection of small-size nodules. Our experiments on the large-scale LIDC dataset demonstrate that the proposed LGDNet with the dual-branch module and attention gate module could significantly improve the nodule detection sensitivity by achieving a final competition performance metric (CPM) score of 89.49%, outperforming the state-of-the-art nodule detection methods, indicating its potential for clinical applications in the early diagnosis of lung diseases.

Deep Learning of Diabetic Retinopathy Classification in Fundus Images

Diabetic retinopathy is an eye disease in diabetic patients due to damage to the small blood vessels in the retina due to high and low blood sugar levels. Accurate detection and classification of Diabetic Retinopathy is an important task in computer-aided diagnosis, especially when planning for diabetic retinopathy surgery. Therefore, this study aims to design an automated model based on deep learning, which helps ophthalmologists detect and classify diabetic retinopathy severity through fundus images. In this work, a deep convolutional neural network (CNN) with transfer learning and fine tunes has been proposed by using pre-trained networks known as Residual Network-50 (ResNet-50). The overall framework of the proposed classification model is divided into three major phases, including pre-processing, training the Resnet-50 network, and classification with evaluation. In the first phase, pre-processing techniques are applied to the APTOS2019 fundus images dataset to find the best features and highlight some fine details of these images. The resnet-50 network was trained in the second phase using the training set and saved the best model obtained that gives high accuracy during the training process. Finally, this saved model has been implemented on the testing dataset for classification DR grades. The proposed model shows good and best classification performance, which was obtained with an accuracy of 98.3%, a precision of 98.4%, an F1-Score of 98.5 % and the recall of 98.4%.

Bayesian uncertainty estimation for detection of long-tailed and unseen conditions in medical images.

Deep supervised learning provides an effective approach for developing robust models for various computer-aided diagnosis tasks. However, there is often an underlying assumption that the frequencies of the samples between the different classes of the training dataset are either similar or balanced. In real-world medical data, the samples of positive classes often occur too infrequently to satisfy this assumption. Thus, there is an unmet need for deep-learning systems that can automatically identify and adapt to the real-world conditions of imbalanced data. We propose a deep Bayesian ensemble learning framework to address the representation learning problem of long-tailed and out-of-distribution (OOD) samples when training from medical images. By estimating the relative uncertainties of the input data, our framework can adapt to imbalanced data for learning generalizable classifiers. We trained and tested our framework on four public medical imaging datasets with various imbalance ratios and imaging modalities across three different learning tasks: semantic medical image segmentation, OOD detection, and in-domain generalization. We compared the performance of our framework with those of state-of-the-art comparator methods. Our proposed framework outperformed the comparator models significantly across all performance metrics (pairwise -test: ) in the semantic segmentation of high-resolution CT and MR images as well as in the detection of OOD samples (), thereby showing significant improvement in handling the associated long-tailed data distribution. The results of the in-domain generalization also indicated that our framework can enhance the prediction of retinal glaucoma, contributing to clinical decision-making processes. Training of the proposed deep Bayesian ensemble learning framework with dynamic Monte-Carlo dropout and a combination of losses yielded the best generalization to unseen samples from imbalanced medical imaging datasets across different learning tasks.

Hybrid-supervised bidirectional transfer networks for computer-aided diagnosis

Medical imaging techniques have been widely used for diagnosis of various diseases. However, the imaging-based diagnosis generally depends on the clinical skill of radiologists. Computer-aided diagnosis (CAD) can help radiologists improve diagnostic accuracy as well as the consistency and reproducibility. Although convolutional neural network (CNN) has shown its feasibility and effectiveness in CAD, it generally suffers from the problem of small sample size when training CAD models. Nowadays, self-supervised learning (SSL) has shown its effectiveness in the field of medical image analysis, especially when there are only limited training samples. However, the backbone of downstream task sometimes cannot be well pre-trained in the conventional SSL framework due to the limitation of the pretext task and fine-tuning mechanism. In this work, an improved SSL framework, named Hybrid-supervised Bidirectional Transfer Networks (HBTN), is proposed to improve the performance of CAD models. Specifically, a novel Gray-Scale Image Mapping (GSIM) task is developed, which still takes the widely used image restoration task in SSL as the pretext task, but further embeds the class label information into it to improve discriminative feature learning of its corresponding network model. The proposed HBTN then integrates two different network architectures, i.e. the image restoration network for the pretext task and the classification network for the downstream task, into a unified hybrid-supervised learning (HSL) framework. It jointly trains both networks and collaboratively transfers the knowledge between each other. Consequently, the performance of downstream network is thus improved. The proposed HBTN is evaluated on two medical image datasets for CAD tasks. The experimental results indicate that HBTN outperforms the conventional SSL algorithms for CAD with limited training samples.

OCIF: automatically learning the optimized clinical information fusion method for computer-aided diagnosis tasks.

In computer-aided diagnosis, the fusion of image features extracted from neural networks and clinical information is crucial to improve diagnostic accuracy. How to integrate low-dimensional clinical information (LDCF) with high-dimensional network features (HDNF) is an urgent problem to be solved. We offer a new network search framework to address this problem, which can provide optimized LDCF fusion and efficient dimensionality reduction in HDNF. OCIF innovatively uses Gaussian process optimization to explore the search space for the number of fully connected (FC) layers, the number of neurons in each FC layer, the activation function, the dropout factor, and whether to add clinical information to each FC layer. Moreover, OCIF employs transfer learning to reduce the training parameter space and improve search efficiency. To evaluate the effectiveness of the proposed OCIF, we utilized three popular end-to-end overall survival (OS) time prediction models to predict the three classes. Our experimental results show that applying OCIF to a classical computer-aided diagnosis neural network can improve classification accuracy. Experiments on the 2020 BRATS dataset prove that OCIF achieves satisfactory performance, with an accuracy of 0.684, precision of 0.735, recall of 0.684, and F1-score of 0.675 on the OS time prediction task. OCIF effectively and creatively combines clinical information and network features, leveraging both clinical information and image features to enhance the accuracy of the final diagnosis. Our experiments demonstrate that the use of OCIF can significantly improve computer-aided diagnosis accuracy, and the approach has the potential to be extended to other medical classification tasks as well.

GA-Net: Ghost convolution adaptive fusion skin lesion segmentation network

Automatic segmentation of skin lesions is a pivotal task in computer-aided diagnosis, playing a crucial role in the early detection and treatment of skin cancer. Despite the existence of numerous deep learning-based segmentation methods, the extraction of lesion features remains inadequate as a result of the segmentation process. Consequently, skin lesion image segmentation continues to face challenges regarding missing detailed information and inaccurate segmentation of the lesion region. In this paper, we propose a ghost convolution adaptive fusion network for skin lesion segmentation. First, the neural network incorporates a ghost module instead of the ordinary convolution layer, generating a rich skin lesion feature map for comprehensive target feature extraction. Subsequently, the network employs an adaptive fusion module and bilateral attention module to connect the encoding and decoding layers, facilitating the integration of shallow and deep network information. Moreover, multi-level output patterns are used for pixel prediction. Layer feature fusion effectively combines output features of different scales, thus improving image segmentation accuracy. The proposed network was extensively evaluated on three publicly available datasets: ISIC2016, ISIC2017, and ISIC2018. The experimental results demonstrated accuracies of 96.42%, 94.07%, and 95.03%, and kappa coefficients of 90.41%, 81.08%, and 86.96%, respectively. The overall performance of our network surpassed that of existing networks. Simulation experiments further revealed that the ghost convolution adaptive fusion network exhibited superior segmentation results for skin lesion images, offering new possibilities for the diagnosis of skin diseases.

A Lie group kernel learning method for medical image classification

Medical image classification is a basic step in medical image analysis and has been an essential task in computer-aided diagnosis. Existing classification methods are proved to be effective in conventional image classification tasks, but they often achieve a suboptimal performance when applied to medical images characterizing by complex nonlinear variation. Aiming at this challenge, this paper proposes a Lie group kernel learning method for medical image classification by combining Lie group theory, kernel functions, SVM and KNN classifiers. The method represents each image with a Lie group feature descriptor constructed from low-level features and builds a SVM classifier from the training images. Geodesic distances between categorical pivots and each testing image are calculated with Lie group kernel functions to select either the SVM or a KNN classifier to do the classification. The proposed method is applied to three medical image datasets and the results demonstrate the efficacy of the method.

Medical image segmentation method based on multi-feature interaction and fusion over cloud computing

Medical image segmentation is a crucial task in computer-aided diagnosis. While deep learning has significantly improved this field, relying solely on local computing power makes it challenging to achieve real-time segmentation results. Furthermore, traditional convolutional neural networks (CNNs) lack the ability to extract global features. To address these issues, this paper proposes a cloud-based medical image segmentation method that leverages multi-feature extraction and interactive fusion. Specifically, this method employs cloud computing to process a large number of medical images and overcome local computing power limitations. It also combines Transformer and CNNs to extract global and local features, respectively, and introduces an interactive fusion attention module to improve segmentation accuracy. The proposed approach is validated on multiple medical image datasets, and experimental results demonstrate its effectiveness and progress.

Optimization System Based on Convolutional Neural Network and Internet of Medical Things for Early Diagnosis of Lung Cancer.

Recently, deep learning and the Internet of Things (IoT) have been widely used in the healthcare monitoring system for decision making. Disease prediction is one of the emerging applications in current practices. In the method described in this paper, lung cancer prediction is implemented using deep learning and IoT, which is a challenging task in computer-aided diagnosis (CAD). Because lung cancer is a dangerous medical disease that must be identified at a higher detection rate, disease-related information is obtained from IoT medical devices and transmitted to the server. The medical data are then processed and classified into two categories, benign and malignant, using a multi-layer CNN (ML-CNN) model. In addition, a particle swarm optimization method is used to improve the learning ability (loss and accuracy). This step uses medical data (CT scan and sensor information) based on the Internet of Medical Things (IoMT). For this purpose, sensor information and image information from IoMT devices and sensors are gathered, and then classification actions are taken. The performance of the proposed technique is compared with well-known existing methods, such as the Support Vector Machine (SVM), probabilistic neural network (PNN), and conventional CNN, in terms of accuracy, precision, sensitivity, specificity, F-score, and computation time. For this purpose, two lung datasets were tested to evaluate the performance: Lung Image Database Consortium (LIDC) and Linear Imaging and Self-Scanning Sensor (LISS) datasets. Compared to alternative methods, the trial outcomes showed that the suggested technique has the potential to help the radiologist make an accurate and efficient early lung cancer diagnosis. The performance of the proposed ML-CNN was analyzed using Python, where the accuracy (2.5-10.5%) was high when compared to the number of instances, precision (2.3-9.5%) was high when compared to the number of instances, sensitivity (2.4-12.5%) was high when compared to several instances, the F-score (2-30%) was high when compared to the number of cases, the error rate (0.7-11.5%) was low compared to the number of cases, and the computation time (170 ms to 400 ms) was low compared to how many cases were computed for the proposed work, including previous known methods. The proposed ML-CNN architecture shows that this technique outperforms previous works.

ESTAN: Enhanced Small Tumor-Aware Network for Breast Ultrasound Image Segmentation.

Breast tumor segmentation is a critical task in computer-aided diagnosis (CAD) systems for breast cancer detection because accurate tumor size, shape, and location are important for further tumor quantification and classification. However, segmenting small tumors in ultrasound images is challenging due to the speckle noise, varying tumor shapes and sizes among patients, and the existence of tumor-like image regions. Recently, deep learning-based approaches have achieved great success in biomedical image analysis, but current state-of-the-art approaches achieve poor performance for segmenting small breast tumors. In this paper, we propose a novel deep neural network architecture, namely the Enhanced Small Tumor-Aware Network (ESTAN), to accurately and robustly segment breast tumors. The Enhanced Small Tumor-Aware Network introduces two encoders to extract and fuse image context information at different scales, and utilizes row-column-wise kernels to adapt to the breast anatomy. We compare ESTAN and nine state-of-the-art approaches using seven quantitative metrics on three public breast ultrasound datasets, i.e., BUSIS, Dataset B, and BUSI. The results demonstrate that the proposed approach achieves the best overall performance and outperforms all other approaches on small tumor segmentation. Specifically, the Dice similarity coefficient (DSC) of ESTAN on the three datasets is 0.92, 0.82, and 0.78, respectively; and the DSC of ESTAN on the three datasets of small tumors is 0.89, 0.80, and 0.81, respectively.

ML-DSVM+: A meta-learning based deep SVM+ for computer-aided diagnosis

Transfer learning (TL) can improve the performance of a single-modal medical imaging-based computer-aided diagnosis (CAD) by transferring knowledge from related imaging modalities. Support vector machine plus (SVM+) is a supervised TL classifier specially designed for TL between the paired data in the source and target domains with shared labels. In this work, a novel deep neural network (DNN) based SVM+ (DSVM+) algorithm is proposed for single-modal imaging-based CAD. DSVM+ integrates the bi-channel DNNs and SVM+ classifier into a unified framework to improve the performance of both feature representation and classification. In particular, a new coupled hinge loss function is developed to conduct bidirectional TL between the source and target domains, which further promotes knowledge transfer together with the feature representation under the guidance of shared labels. To alleviate the overfitting caused by the increased parameters in DNNs for limited training samples, the meta-learning based DSVM+ (ML-DSVM+) is further developed, which designs randomly selecting samples from the training data instead of other CAD tasks for meta-tasks. This sampling strategy also can avoid the issue of class imbalance. ML-DSVM+ is evaluated on three medical imaging datasets. It achieves the best results of 88.26±1.40%, 90.45±5.00%, and 87.63±5.56% on accuracy, sensitivity and specificity, respectively, on the Bimodal Breast Ultrasound Image dataset, 90.00±1.05%, 72.55±3.87%, and 96.40±2.26% of the corresponding indices on the Alzheimer's Disease Neuroimaging Initiative dataset, and 85.76±3.12% of classification accuracy, 88.73±7.22% of sensitivity, and 82.60±1.56% of specificity for the Autism Brain Imaging Data Exchange dataset.

SWTRU: Star-shaped Window Transformer Reinforced U-Net for medical image segmentation

In the last decade, deep neural networks have been widely applied to medical image segmentation, achieving good results in computer-aided diagnosis tasks etc. However, the task of segmenting highly complex, low-contrast images of organs and tissues with high accuracy still faces great challenges. To better address this challenge, this paper proposes a novel model SWTRU (Star-shaped Window Transformer Reinforced U-Net) by combining the U-Net network which plays well in the image segmentation field, and the Transformer which possesses a powerful ability to capture global contexts. Unlike the previous methods that import the Transformer into U-Net, an improved Star-shaped Window Transformer is introduced into the decoder of the SWTRU to enhance the decision-making capability of the whole method. The SWTRU uses a redesigned multi-scale skip-connection model, which retains the inductive bias of the original FCN structure for images while obtaining fine-grained features and coarse-grained semantic information. Our method also presents the FFIM (Filtering Feature Integration Mechanism) to integration and dimensionality reduction of the fused multi-layered features, which reduces the computation. Our SWTRU yields 0.972 DICE on CHLISC for liver and tumor segmentation, 0.897 DICE on LGG for glioma segmentation, and 0.904 DICE on ISIC2018 for skin diseases’ segmentation, achieves substantial improvements over the current SoTA across 9 different medical image segment methods. SWTRU can combine feature mapping from different scales, high-level semantics, and global contextual relationships, this architecture is effective in the medical image segmentation. The experimental findings indicate that SWTRU produces superior performance on the medical image segmentation tasks.

Detection of mitotic HEp-2 cell images: role of feature representation and classification framework under class skew.

We propose and analyze a framework to detect and identify the mitotic type staining patterns among different non-mitotic (interphase) patterns on HEp-2 cell substrate specimen images. This is considered as a principal task in computer-aided diagnosis (CAD) of the autoimmune disorders. Due to the rare appearance of mitotic patterns in whole slide/specimen images, the sample skew between mitotic and non-mitotic patterns is an important consideration.We suggest to apply some effective samples skew balancing strategies for the task of classification between mitotic v/s interphase patterns. Another aspect of this study is to consider the morphology and texture-based differences between both the classes that can be incorporated through effective morphology and texture-based descriptors, including the Gabor and LM (Leung-Malik) filter banks and also through some contemporary filter banks derived from convolutional neural networks (CNN).The proposed framework is evaluated on a public dataset and we demonstrate good performance (0.99 or 1 Matthews correlation coefficient (MCC) in many cases), across various experiments. The study also presents a comparison between hand-engineered and CNN-based feature representation, along with the comparisons with state-of-the-art approaches. Hence, the framework proves to be a good solution for the mentioned skewed classification problem.

Judgment of benign and early malignant colorectal tumors from ultrasound images with deep multi-View fusion

Background and objective: Colorectal cancer (CRC) is currently one of the main cancers world-wide, with a high incidence in the elderly. In the diagnosis of CRC, endorectal ultrasound plays an important role in judging benign and early malignant tumors. However, malignant tumors in the early-stage are not easy to identify visually and experts usually seek help from multi-view images, which increases the workload and also exists a certain probability of misdiagnosis. In recent years, with the widespread use of deep learning methods in the analysis of medical images, it becomes necessary to design an effective computer-aided diagnosis (CAD) system of CRC based on multi-view endorectal ultrasound images. Method: In this study, we proposed a CAD system for judging benign and early malignant colorectal tumors, and constructed the first multi-view ultrasound image dataset of CRC to validate our algorithm. Our system is an end-to-end model based on a deep neural network (DNN) which includes a feature extraction module based on dense blocks, a multi-view fusion module, and a Multi-Layer Perception-based classifier. A center loss was used for the first time in CAD tasks, to optimize our model. Result: On the constructed dataset, the proposed system surpasses expert diagnosis in accuracy, sensitivity, specificity, and F1-score. Compared with the popular deep classification networks and other CAD methods, the algorithm has reached the best performance. Comparative experiments using different feature extraction methods, different view fusion strategies, and different classifiers verify the effectiveness of each part of the algorithm. Conclusion: We propose a CAD system for judging benign and early malignant colorectal tumors based on DNN, which combines information of ultrasound images from different views for comprehension. On the first CRC multi-view ultrasound image dataset which we constructed, our method outperforms expert diagnosis results and all other methods, and the effectiveness of each part of the system has been verified. Our system has application value in future medical practice on early diagnosis of CRC.

Cloud-Based Lung Tumor Detection and Stage Classification Using Deep Learning Techniques.

Artificial intelligence (AI), Internet of Things (IoT), and the cloud computing have recently become widely used in the healthcare sector, which aid in better decision-making for a radiologist. PET imaging or positron emission tomography is one of the most reliable approaches for a radiologist to diagnosing many cancers, including lung tumor. In this work, we proposed stage classification of lung tumor which is a more challenging task in computer-aided diagnosis. As a result, a modified computer-aided diagnosis is being considered as a way to reduce the heavy workloads and second opinion to radiologists. In this paper, we present a strategy for classifying and validating different stages of lung tumor progression, as well as a deep neural model and data collection using cloud system for categorizing phases of pulmonary illness. The proposed system presents a Cloud-based Lung Tumor Detector and Stage Classifier (Cloud-LTDSC) as a hybrid technique for PET/CT images. The proposed Cloud-LTDSC initially developed the active contour model as lung tumor segmentation, and multilayer convolutional neural network (M-CNN) for classifying different stages of lung cancer has been modelled and validated with standard benchmark images. The performance of the presented technique is evaluated using a benchmark image LIDC-IDRI dataset of 50 low doses and also utilized the lung CT DICOM images. Compared with existing techniques in the literature, our proposed method achieved good result for the performance metrics accuracy, recall, and precision evaluated. Under numerous aspects, our proposed approach produces superior outcomes on all of the applied dataset images. Furthermore, the experimental result achieves an average lung tumor stage classification accuracy of 97%-99.1% and an average of 98.6% which is significantly higher than the other existing techniques.

Classification of microcalcification clusters in digital breast tomosynthesis using ensemble convolutional neural network

BackgroundThe classification of benign and malignant microcalcification clusters (MCs) is an important task for computer-aided diagnosis (CAD) of digital breast tomosynthesis (DBT) images. Influenced by imaging method, DBT has the characteristic of anisotropic resolution, in which the resolution of intra-slice and inter-slice is quite different. In addition, the sharpness of MCs in different slices of DBT is quite different, among which the clearest slice is called focus slice. These characteristics limit the performance of CAD algorithms based on standard 3D convolution neural network (CNN).MethodsTo make full use of the characteristics of the DBT, we proposed a new ensemble CNN, which consists of the 2D ResNet34 and the anisotropic 3D ResNet to extract the 2D focus slice features and 3D contextual features of MCs, respectively. Moreover, the anisotropic 3D convolution is used to build 3D ResNet to avoid the influence of DBT anisotropy.ResultsThe proposed method was evaluated on 495 MCs in DBT images of 275 patients, which are collected from our collaborative hospital. The area under the curve (AUC) of receiver operating characteristic (ROC) and accuracy of classifying benign and malignant MCs using decision-level ensemble strategy were 0.8837 and 82.00%, which were significantly higher than the experimental results of 2D ResNet34 (AUC: 0.8264, ACC: 76.00%) and anisotropic 3D ResNet (AUC: 0.8455, ACC: 76.00%). Compared with the results of 3D features classification in the radiomics, the AUC of the deep learning method with decision-level ensemble strategy was improved by 0.0435, and the F1 score was improved from 79.37 to 85.71%. More importantly, the sensitivity increased from 78.13 to 84.38%, and the specificity increased from 66.67 to 77.78%, which effectively reduced the false positives of diagnosisConclusionThe results fully prove that the ensemble CNN can effectively integrate 2D features and 3D features, improve the classification performance of benign and malignant MCs in DBT, and reduce the false positives.

SESV: Accurate Medical Image Segmentation by Predicting and Correcting Errors.

Medical image segmentation is an essential task in computer-aided diagnosis. Despite their prevalence and success, deep convolutional neural networks (DCNNs) still need to be improved to produce accurate and robust enough segmentation results for clinical use. In this paper, we propose a novel and generic framework called Segmentation-Emendation-reSegmentation-Verification (SESV) to improve the accuracy of existing DCNNs in medical image segmentation, instead of designing a more accurate segmentation model. Our idea is to predict the segmentation errors produced by an existing model and then correct them. Since predicting segmentation errors is challenging, we design two ways to tolerate the mistakes in the error prediction. First, rather than using a predicted segmentation error map to correct the segmentation mask directly, we only treat the error map as the prior that indicates the locations where segmentation errors are prone to occur, and then concatenate the error map with the image and segmentation mask as the input of a re-segmentation network. Second, we introduce a verification network to determine whether to accept or reject the refined mask produced by the re-segmentation network on a region-by-region basis. The experimental results on the CRAG, ISIC, and IDRiD datasets suggest that using our SESV framework can improve the accuracy of DeepLabv3+ substantially and achieve advanced performance in the segmentation of gland cells, skin lesions, and retinal microaneurysms. Consistent conclusions can also be drawn when using PSPNet, U-Net, and FPN as the segmentation network, respectively. Therefore, our SESV framework is capable of improving the accuracy of different DCNNs on different medical image segmentation tasks.

FRNet: an end-to-end feature refinement neural network for medical image segmentation

Medical image segmentation is a crucial but challenging task for computer-aided diagnosis. In recent years, fully convolutional network-based methods have been widely applied to medical image segmentation. U-shape-based approaches are one of the most successful structures in this medical field. However, the consecutive down-sampling operations in the encoder lead to the loss of spatial information, which is important for medical image segmentation. In this paper, we present a novel lightweight end-to-end feature refinement network (FRNet) to address this issue. The structure of our model is simple and efficient. Specifically, the network adopts an encoder-decoder network as backbone, where two additional paths, spatial refinement path and semantic refinement path, are applied on the encoder and decoder, respectively, to improve the detailed representation ability and discriminative ability of our model. In addition, we introduce a feature adaptive fusion block (FAF block) that effectively combines features of different depths. The proposed FRNet can be trained in an end-to-end way. We have evaluated our method on three different medical image segmentation tasks. Experimental results show that FRNet has better performance than the state-of-the-art approaches. It achieves a high average accuracy without any post-processing of 0.968 and 0.936 for blood vessel segmentation and skin lesion segmentation, respectively. We further demonstrate that our method can be easily applied to other network structures to improve their performance.