Development of automatic glioma brain tumor detection system using deep convolutional neural networks

We have developed six convolutional neural network (CNN) models for finding optimal brain tumor detection system on high-grade glioma and low-grade glioma lesions from voluminous magnetic resonance imaging human brain scans. Glioma is the most common form of brain tumor. The models are constructed based on the different combinations and settings of hyperparameters with conventional CNN architecture. The six models are two layers with five epochs, five layers with dropout, five layers with stopping criteria (FLSC), FLSC and dropout (FLSCD), FLSC and batch normalization (FLSCBN), and FLSCBN and dropout. The models were trained and tested with BraTS2013 and whole brain atlas data sets. Among them, FLSCBN model yielded the best classification results for brain tumor detection. Experimental results revealed that our deep learning approach was better than the conventional state-of-art methods.     

Performance analysis of computer aided brain tumor detection system using ANFIS classifier

The abrupt changes in brain cells due to the environmental effects or genetic disorders leads to form the abnormal lesions in brain. These abnormal lesions are combined as mass and known as tumor. The detection of these tumor cells in brain image is a complex task due to the similarities between normal cells and tumor cells. In this paper, an automated brain tumor detection and segmentation methodology is proposed. The proposed method consists of feature extraction, classification and segmentation. In this paper, Grey Level Co‐Occurrence Matrix (GLCM), Discrete Wavelet Transform (DWT) and Law's texture features are used as features. These features are fed to Adaptive Neuro Fuzzy Inference System (ANFIS) classifier as input pattern, which classifies the brain image. Morphological operations are now applied on the classified abnormal brain image to segment the tumor regions. The proposed system achieves 95.07% of sensitivity, 99.84% of specificity and 99.80% of accuracy for tumor segmentation.     

基于CNN的点云图像融合目标检测

针对自动驾驶场景中目标检测存在尺度变化、光照变化和缺少距离信息等问题,提出一种极具鲁棒性的多模态数据融合目标检测方法,其主要思想是利用激光雷达提供的深度信息作为附加的特征来训练卷积神经网络(CNN)。首先利用滑动窗对输入数据进行切分匹配网络输入,然后采用两个CNN特征提取器提取RGB图像和点云深度图的特征,将其级联得到融合后的特征图,送入目标检测网络进行候选框的位置回归与分类,最后进行非极大值抑制(NMS)处理输出检测结果,包含目标的位置、类别、置信度和距离信息。在KITTI数据集上的实验结果表明,本文方法通过多模态数据的优势互补提高了在不同光照场景下的检测鲁棒性,附加滑动窗处理改善了小目标的检测效果。对比其他多种检测方法,本文方法具有检测精度与检测速度上的综合优势。     

Computer aided automated detection and classification of brain tumors using CANFIS classification method

The development of abnormal cells in human brain leads to the formation of tumors. This article proposes an efficient approach for brain tumor detection and segmentation using image fusion and co-active adaptive neuro fuzzy inference system (CANFIS) classification method. The brain MRI images are fused and the dual tree complex wavelet transform is applied on the fused image. Then, the statistical features, local ternary pattern features and gray level co-occurrence matrix features. These extracted features are classified using CANFIS classification approach for the classification of source brain MRI image into either normal or abnormal. Further, morphological operations are applied on the abnormal brain MRI image for segmenting the tumor regions. The proposed methodology is evaluated with respect to the performance metrics sensitivity, specificity, positive predictive value, negative predictive value, tumor segmentation accuracy with detection rate. The proposed image fusion based brain tumor detection and classification methodology stated in this article achieves 96.5% of average sensitivity, 97.7% of average specificity, 87.6% of positive predictive value, 96.6% of negative predictive value, and 98.8% of average accuracy.     

An efficient and automatic glioblastoma brain tumor detection using shift‐invariant shearlet transform and neural networks

The detection and segmentation of tumor region in brain image is a critical task due to the similarity between abnormal and normal region. In this article, a computer‐aided automatic detection and segmentation of brain tumor is proposed. The proposed system consists of enhancement, transformation, feature extraction, and classification. The shift‐invariant shearlet transform (SIST) is used to enhance the brain image. Further, nonsubsampled contourlet transform (NSCT) is used as multiresolution transform which transforms the spatial domain enhanced image into multiresolution image. The texture features from grey level co‐occurrence matrix (GLCM), Gabor, and discrete wavelet transform (DWT) are extracted with the approximate subband of the NSCT transformed image. These extracted features are trained and classified into either normal or glioblastoma brain image using feed forward back propagation neural networks. Further, K‐means clustering algorithm is used to segment the tumor region in classified glioblastoma brain image. The proposed method achieves 89.7% of sensitivity, 99.9% of specificity, and 99.8% of accuracy.     

Content‐based gastric image retrieval using convolutional neural networks

The endoscopy procedure has demonstrated great efficiency in detecting stomach lesions, with extensive numbers of endoscope images produced globally each day. The content‐based gastric image retrieval (CBGIR) system has demonstrated substantial potential in gastric image analysis. Gastric precancerous diseases (GPD) have higher prevalence in gastric cancer patients. Thus, effective intervention is crucial at the GPD stage. In this paper, a CBGIR method is proposed using a modified ResNet‐18 to generate binary hash codes for a rapid and accurate image retrieval process. We tested several popular models (AlexNet, VGGNet and ResNet), with ResNet‐18 determined as the optimum option. Our proposed method was valued using a GPD data set, resulting in a classification accuracy of 96.21 ± 0.66% and a mean average precision of 0.927 ± 0.006 , outperforming other state‐of‐art conventional methods. Furthermore, we constructed a Gastric‐Map (GM) based on feature representations in order to visualize the retrieval results. This work has great auxiliary significance for endoscopists in terms of understanding the typical GPD characteristics and improving aided diagnosis.     

Cell-type based semantic segmentation of histopathological images using deep convolutional neural networks

Histopathological whole-slide image (WSI) analysis is still one of the most important ways to identify regions of cancer risk. For cancer in which early diagnosis is vital, pathologists are at the center of the decision-making process. Thanks to the widespread use of digital pathology and the development of artificial intelligence methods, automatic histopathological image analysis methods help pathologists in their decision-making process. In this process, rather than producing labels for whole-slide image patches, semantic segmentation is very useful, which facilitates the pathologists’ interpretation. In this study, automatic semantic segmentation based on cell type is proposed for the first time in the literature using novel deep convolutional networks structure (DCNN). We presents semantic information on four classes, including white areas in the whole-slide image, tissue without cells, tissue with normal cells and tissue with cancerous cells. This visual information presented to the pathologist is an easy-to-understand picture of the status of the cells and their implications for the spread of cancerous cells. A new DCNN architecture is created, inspired by the residual network and deconvolution network architecture. Our network is trained end-to-end manner with histopathological image patches for cell structures to be more discriminative. The proposed method not only produces more successful results than other state-of-art semantic segmentation algorithms with 9.2% training error and 88.89% F-score for test, but also has the most important advantage in that it has the ability to generate automatic information about the cancer and also provides information that pathologists can quickly interpret.     

DeepESC网络的环境声分类方法研究

为进一步提升环境声分类的识别率,提出了一种仿深度隐藏身份特征（Deep Hidden Identity Feature,DeepID）网络连接方式的卷积神经网络——深度环境声分类网络（Deep Environment Sound Classification,DeepESC）。DeepESC网络共有六层——三层卷积层、两层全连层以及一层聚合层,为使网络在自动抽取高层次特征的同时能有效地兼顾低层次特征,网络将三层卷积层的输出聚合为一层,该层充分包含不同层次的特征,提升了卷积神经网络的特征表达能力。ESC-10和ESC-50数据集上的仿真结果表明：在相同的识别框架下,与随机森林分类器相比,本文网络识别率分别平均提升了7.6%和22.4%,与传统的卷积神经网络相比,识别率分别平均提升4%和2%,仿真实验验证了本文分类器的有效性。     

Defect pattern recognition on wafers using convolutional neural networks

In semiconductor manufacturing, wafer testing is performed to ensure the performance of each product after wafer fabrication. The wafer map is used to visualize the color-coded wafer test results based on the locations. The defects on the wafer map may be randomly distributed or form clustered patterns. The various clustered defect patterns are usually caused by assignable faults. The identification of the patterns is thus important to provide valuable hints for the root causes diagnosis. Solving the problems helps improve the manufacturing processes and reduce costs. In this study, we present a novel convolutional neural network (CNN)–based method to automatically recognize the defect pattern on wafer maps. Our method uses polar mapping before the training of CNN to transform the circular wafer map into a matrix, which can be processed within CNN architecture. This procedure also reduces the input size and solves variations in wafer sizes and die sizes. To eliminate the effects of rotation, we apply data augmentation in the training of CNN. Experiments using the real-world dataset prove the effectiveness and superiority of our method.     

Detection and diagnosis of meningioma brain tumor using ANFIS classifier

Tumors are formed in brain due to the uncontrolled development of cells. These tumors can be cured if it is timely detected and by proper medication. This article proposes a computer‐aided automatic detection and diagnosis of meningioma brain tumors in brain images using Adaptive Neuro Fuzzy Inference System (ANFIS) classifier. The proposed system consists of feature extraction, classification, and segmentation and diagnosis sections. In this article, Grey level Co‐occurrence Matric (GLCM) and Grid features are extracted from the brain image and these features are classified using ANFIS classifier into normal or abnormal. Then, morphological operations are used to segment the abnormal regions in brain image. Based on the location of these abnormal regions in brain tissues, the segmented tumor regions are diagnosed.     

A computer‐aided approach for meningioma brain tumor detection using CANFIS classifier

Abnormal growth of cells in brain leads to the formation of tumors in brain. The earlier detection of the tumors in brain will save the life of the patients. Hence, this article proposes a computer‐aided fully automatic methodology for brain tumor detection using Co‐Active Adaptive Neuro Fuzzy Inference System (CANFIS) classifier. The internal region of the brain image is enhanced using image normalization technique and further contourlet transform is applied on the enhanced brain image for the decomposition with different scales. The grey level and heuristic features are extracted from the decomposed coefficients and these features are trained and classified using CANFIS classifier. The performance of the proposed brain tumor detection is analyzed in terms of classification accuracy, sensitivity, specificity, and segmentation accuracy.     

Multiview convolutional neural networks for lung nodule classification

To find a better way to screen early lung cancer, motivated by the great success of deep learning, we empirically investigate the challenge of classifying lung nodules in computed tomography (CT) in an end‐to‐end manner. Multi‐view convolutional neural networks (MV‐CNN) are proposed in this article for lung nodule classification. Unlike the traditional CNNs, a MV‐CNN takes multiple views of each entered nodule. We carry out a binary classification (benign and malignant) and a ternary classification (benign, primary malignant, and metastatic malignant) using the Lung Image Database Consortium and Image Database Resource Initiative database. The results show that, for binary or ternary classifications, the multiview strategy produces higher accuracy than the single view method, even for cases that are over‐fitted. Our model achieves an error rate of 5.41 and 13.91% for binary and ternary classifications, respectively. Finally, the receiver operating characteristic curve and t‐distributed stochastic neighbor embedding algorithm are used to analyze the models. The results reveal that the deep features learned by the model proposed in this article have a higher separability than features from the image space and the multiview strategies; therefore, researchers can get better representation. © 2017 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 27, 12–22, 2017     

Pre-trained convolutional neural networks as feature extractors for diagnosis of breast cancer using histopathology

Several researchers are trying to develop different computer-aided diagnosis system for breast cancer employing machine learning (ML) methods. The inputs to these ML algorithms are labeled histopathological images which have complex visual patterns. So, it is difficult to identify quality features for cancer diagnosis. The pre-trained Convolutional Neural Networks (CNNs) have recently emerged as an unsupervised feature extractor. However, a limited investigation has been done for breast cancer recognition using histopathology images with CNN as a feature extractor. This work investigates ten different pre-trained CNNs for extracting the features from breast cancer histopathology images. The breast cancer histopathological images are obtained from publicly available BreakHis dataset. The classification models for the different feature sets, which are obtained using different pre-trained CNNs in consideration, are developed using a linear support vector machine. The proposed method outperforms the other state of art methods for cancer detection, which can be observed from the results obtained.     

Smart connected electronic gastroscope system for gastric cancer screening using multi-column convolutional neural networks

Gastroscopy is a widely adopted method for gastric cancer screening and early diagnosis. Clinical studies show that it can effectively prolong patient life and maximise therapeutic effect. However, it is difficult for doctors to identify and detect lesions in real time, which manifests as the major challenge in gastroscopy. In this paper, we propose SCEG, a smart connected electronic gastroscopy system that performs dynamic cancer screening in gastroscopy. By integrating electronic gastroscopy with cloud-based medical image analysis service, we develop an AdaBoost-based multi-column convolutional neural network (MCNN) for enhancing gastric cancer screening. Experimental results show that the proposed MCNN approach significantly outperforms other competing approaches.     

Deep convolutional neural networks for eigenvalue problems in mechanics

We show that deep convolutional neural networks (CNNs) can massively outperform traditional densely connected neural networks (NNs) (both deep or shallow) in predicting eigenvalue problems in mechanics. In this sense, we strike out in a new direction in mechanics computations with strongly predictive NNs whose success depends not only on architectures being deep but also being fundamentally different from the widely used to date. We consider a model problem: predicting the eigenvalues of one-dimensional (1D) and two-dimensional (2D) phononic crystals. For the 1D case, the optimal CNN architecture reaches 98% accuracy level on unseen data when trained with just 20 000 samples, compared to 85% accuracy even with 100 000 samples for the typical network of choice in mechanics research. We show that, with relatively high data efficiency, CNNs have the capability to generalize well and automatically learn deep symmetry operations, easily extending to higher dimensions and our 2D case. Most importantly, we show how CNNs can naturally represent mechanical material tensors, with its convolution kernels serving as local receptive fields, which is a natural representation of mechanical response. Strategies proposed are applicable to other mechanics' problems and may, in the future, be used to sidestep cumbersome algorithms with purely data-driven approaches based upon modern deep architectures.     

Fusion of MRI and CT images using guided image filter and image statistics

In medical imaging using different modalities such as MRI and CT, complementary information of a targeted organ will be captured. All the necessary information from these two modalities has to be integrated into a single image for better diagnosis and treatment of a patient. Image fusion is a process of combining useful or complementary information from multiple images into a single image. In this article, we present a new weighted average fusion algorithm to fuse MRI and CT images of a brain based on guided image filter and the image statistics. The proposed algorithm is as follows: detail layers are extracted from each source image by using guided image filter. Weights corresponding to each source image are calculated from the detail layers with help of image statistics. Then a weighted average fusion strategy is implemented to integrate source image information into a single image. Fusion performance is assessed both qualitatively and quantitatively. Proposed method is compared with the traditional and recent image fusion methods. Results showed that our algorithm yields superior performance.     

An Efficient Image Analysis Framework for the Classification of Glioma Brain Images Using CNN Approach

The identification of brain tumors is multifarious work for the separation of the similar intensity pixels from their surrounding neighbours. The detection of tumors is performed with the help of automatic computing technique as presented in the proposed work. The non-active cells in brain region are known to be benign and they will never cause the death of the patient. These non-active cells follow a uniform pattern in brain and have lower density than the surrounding pixels. The Magnetic Resonance (MR) image contrast is improved by the cost map construction technique. The deep learning algorithm for differentiating the normal brain MRI images from glioma cases is implemented in the proposed method. This technique permits to extract the linear features from the brain MR image and glioma tumors are detected based on these extracted features. Using k-mean clustering algorithm the tumor regions in glioma are classified. The proposed algorithm provides high sensitivity, specificity and tumor segmentation accuracy.     

基于卷积网络的目标跟踪应用研究

本文针对目标跟踪应用,提出了基于Siamese-FC跟踪网络的改进卷积网络Siamese-MF,意在更进一步提升跟踪速度和准确性,满足目标跟踪的工程应用需求。对于跟踪网络,考虑速度和精度的权衡,减少计算量,增加卷积特征的感受野是改进跟踪网络的速度和精度的方向。在卷积网络结构上面进行改进结构创新,改进主要集中为两点:1)引入特征融合,丰富特征;2)引入空洞卷积,减少计算量的同时增强感受野。Siamese-MF算法实现了对于复杂场景目标的实时准确跟踪,在公开数据集OTB上测试速度达到平均76 f/s,跟踪成功率的均值达到0.44,而跟踪稳定性的均值达到0.61,实时性、准确性和稳定性均提升,满足目标实时跟踪应用。     

An intelligent mining system for diagnosing medical images using combined texture‐histogram features

The aim of this article is to design an expert system for medical image diagnosis. We propose a method based on association rule mining combined with classification technique to enhance the diagnosis of medical images. This system classifies the images into two categories namely benign and malignant. In the proposed work, association rules are extracted for the selected features using an algorithm called AprioriTidImage, which is an improved version of Apriori algorithm. Then, a new associative classifier CLASS_Hiconst ( CL assifier based on ASS ociation rules with Hi gh Con fidence and S uppor t ) is modeled and used to diagnose the medical images. The performance of our approach is compared with two different classifiers Fuzzy‐SVM and multilayer back propagation neural network (MLPNN) in terms of classifier efficiency with sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. The experimental result shows 96% accuracy, 97% sensitivity, and 96% specificity and proves that association rule based classifier is a powerful tool in assisting the diagnosing process. © 2013 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 23, 194–203, 2013     

多路径递归网络结构的单帧图像超分辨率重建

卷积神经网络在单帧图像超分辨率重建任务中取得了巨大成功,但是其重建模型多是基于单链结构,层间联系较弱且不能充分利用网络提取的分层特征。针对这些问题,本文设计了一种多路径递归的网络结构(MRCN)。通过使用多路径结构来加强层之间的联系,实现特征的有效利用并且提取丰富的高频成分,同时使用递归结构降低训练难度。此外,通过引入特征融合的操作使得在重建的过程中可以充分利用各层提取的特征,并且自适应的选择有效特征。在常用的基准测试集上进行了大量实验表明,MRCN比现有的方法在重建效果上具有明显提升。