In this study, we proposed a novel convolutional neural network (CNN) architecture for classification of benign and malignant breast cancer (BC) in histological images. To improve the delivery and use of feature information, we chose the DenseNet as the basic building block and interleaved it with the squeeze-and-excitation (SENet) module. We conducted extensive experiments with the proposed framework by using the public domain BreakHis dataset and demonstrated that the proposed framework can produce significantly improved accuracy in BC classification, compared with the state-of-the-art CNN methods reported in the literature.
Cervical cancer is a worldwide public health problem with a high rate of illness and mortality among women. In this study, we proposed a novel framework based on Faster RCNN-FPN architecture for the detection of abnormal cervical cells in cytology images from a cancer screening test. We extended the Faster RCNN-FPN model by infusing deformable convolution layers into the feature pyramid network (FPN) to improve scalability. Furthermore, we introduced a global contextual aware module alongside the Region Proposal Network (RPN) to enhance the spatial correlation between the background and the foreground. Extensive experimentations with the proposed deformable and global context aware (DGCA) RCNN were carried out using the cervical image dataset of “Digital Human Body” Vision Challenge from the Alibaba Cloud TianChi Company. Performance evaluation based on the mean average precision (mAP) and receiver operating characteristic (ROC) curve has demonstrated considerable advantages of the proposed framework. Particularly, when combined with tagging of the negative image samples using traditional computer-vision techniques, 6–9% increase in mAP has been achieved. The proposed DGCA-RCNN model has potential to become a clinically useful AI tool for automated detection of cervical cancer cells in whole slide images of Pap smear.
RGB-Infrared person re-identification (RGB-IR Re-ID) is a cross-modality matching problem with promising applications in the dark environment. Most existing works use Euclidean metric based constraints to resolve the discrepancy between features of different modalities. However, these methods are incapable of learning angularly discriminative feature embedding because Euclidean distance cannot measure the included angle between embedding vectors effectively. As an angularly discriminative feature space is important for classifying the human images based on their embedding vectors, in this paper, we propose a novel ranking loss function, named Bi-directional Exponential Angular Triplet Loss, to help learn an angularly separable common feature space by explicitly constraining the included angles between embedding vectors. Moreover, to help stabilize and learn the magnitudes of embedding vectors, we adopt a common space batch normalization layer. Quantitative experiments on the SYSU-MM01 and RegDB dataset support our analysis. On SYSU-MM01 dataset, the performance is improved from 7.40% / 11.46% to 38.57% / 38.61% for rank-1 accuracy / mAP compared with the baseline. The proposed method can be generalized to the task of single-modality Re-ID and improves the rank-1 accuracy / mAP from 92.0% / 81.7% to 94.7% / 86.6% on the Market-1501 dataset, from 82.6% / 70.6% to 87.6% / 77.1% on the DukeMTMC-reID dataset.
Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is an illness characterized by a diverse range of debilitating symptoms including autonomic, immunologic, and cognitive dysfunction. Although neurological and cognitive aberrations have been consistently reported, relatively little is known regarding the regional cerebral blood flow (rCBF) in ME/CFS. In this study, we studied a cohort of 31 ME/CSF patients (average age: 42.8 ± 13.5 years) and 48 healthy controls (average age: 42.9 ± 12.0 years) using the pseudo-continuous arterial spin labeling (PCASL) technique on a whole-body clinical 3T MRI scanner. Besides routine clinical MRI, the protocol included a session of over 8 min-long rCBF measurement. The differences in the rCBF between the ME/CSF patients and healthy controls were statistically assessed with voxel-wise and AAL ROI-based two-sample t-tests. Linear regression analysis was also performed on the rCBF data by using the symptom severity score as the main regressor. In comparison with the healthy controls, the patient group showed significant hypoperfusion (uncorrected voxel wise p ≤ 0.001, FWE p ≤ 0.01) in several brain regions of the limbic system, including the anterior cingulate cortex, putamen, pallidum, and anterior ventral insular area. For the ME/CFS patients, the overall symptom severity score at rest was significantly associated with a reduced rCBF in the anterior cingulate cortex. The results of this study show that brain blood flow abnormalities in the limbic system may contribute to ME/CFS pathogenesis.
According to the newly proposed nested MIMO (Multiple-Input Multiple-Input Multiple Output Multiple Array) array design method, we propose to replace the traditional nested array into an optimizing nested array, ie, to optimizing nested MIMO array design. It not only retains the original advantage of nested MIMO array design closed expression with array element position and degree of freedom(DOF), but also greatly improves the array aperture and DOF. Optimizing nested MIMO array firstly uses the optimizing nested array as the transmitting and receiving arrays, and then make the difference set processing for the coarray of MIMO array (coarray, CA). By properly designing the array spacing of the transmitting and receiving arrays, we can obtain a non-porous difference array. When the total number of array elements is given, by analyzing the characteristics of the array structure, the best array element number of the transmitting and receiving arrays can be obtained. Simulation experiments show that compared with the nested MIMO array design, the proposed method can effectively expand the array aperture, increase the DOF, and increase the DOA estimation accuracy of the MIMO radar without increasing the number of actual array elements.
Hybrid Multiple-Input-Multiple-Output(MIMO) phased array radar, or Phased MIMO radar, is recognized for its capacity to provide trade-off between transmit coherent gain and waveform diversity gain. Such trade-off can be flexibly implemented by changing subarray partition schemes of the system, which in turn enhances system performance. In this paper, a center-spanned subarray configuration is proposed with detailed steps to develop: First, the center element of the whole transmit array is chosen. Next, the surrounding region of center element is determined as the center subarray for subsequent spanning. Finally, the subarray partition is implemented through spanning subarrays around the center subarray. Both theoretical derivations and simulation results reveal that the hybrid MIMO phased array radar with the proposed center-spanned subarray configuration has lower sidelobe beampattern and higher Signal to Interference Noise Ratio(SINR) than that of equally or unequally overlapped subarray partitions. INDEX TERMS Beampattern, center-spanned sub-arrays, phased MIMO radar, SINR.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.