A multi-resolution approach for spinal metastasis detection using deep Siamese neural networks

With the increase in the amount of remote sensing images (RSIs), deep learning (DL) has been used to the change detection (CD) task in remote sensing field and achieved good results. However, most existing methods do not take full advantage of the temporal dependence of the multi-temporal images. In this paper, we propose a novel method for CD, namely SNN-LSTM (a deep Siamese neural network (SNN) with convolutional Long Short-Term Memory (ConvLSTM) and channel attention module (CAM)), especially for capturing and representing spatial-temporal information effectively. It mainly contains three parts. First, a network based on Siamese convolutional architecture is designed to extract multi-level features. Then, a ConvLSTM block is introduced to further obtain time dependency of multi-temporal RSIs, and spatial information is also extracted simultaneously. Finally, CAM blocks are used to refine the extracted multi-level features, enhance the feature of changes, and eventually generate change map. The experiments are conducted on LEVIR-CD dataset, both visual results and quantitative assessment prove that the proposed method outperforms several state-of-the-art methods.

Attention shake siamese network with auxiliary relocation branch for visual object tracking

Siamese network is compute-intensive learning model with growing applicability in a wide range of domains. However, state-of-art deep neural network (DNN) accelerators would not work efficiently for Siamese network, as their designs do not account for the algorithm properties of Siamese network. In this paper, we propose a Siamese network accelerator called SNA, the first Simultaneous Multi-Threading (SMT) hardware architecture to perform Siamese network inference with high performance and energy efficiency. We devise an adaptive inter-model computing resource partition and flexible on-chip buffer management mechanism based on the model parallelism and SMT design philosophy. Our architecture is implemented in Verilog and synthesized in a 65nm technology using Synopsys design tools. We also evaluate it with several typical Siamese networks. Compared to the state-of-art accelerator, on average, the SNA architecture offers 2.1x speedup and 1.48x energy reduction.

Background: Approximately 8% of breast cancers originate within the central ducts near the nipple-areolar complex(NAC), which revealed higher rates of occult involvement than previously thought. Precise preoperative imaging to assess NAC and subjacent tumor involvement has become critical to identify candidates for nipple-sparing mastectomy(NSM) while ensuring oncological safety. The aim of our study is to determine whether MRI could be used for the detection of subtle neoplastic infiltration. Method: Electronic databases including PubMed, Scopus, Web of Science, and EMBASE were systematically searched to identify all relevant studies on diagnostic performance of Magnetic Resonance Imaging(MRI) in malignant extension to the nipple published before 2024. The keywords include MRI, NAC, breast cancer, NSM. True Positive(TP), True Negative(TN), False Positive(FP), False Negative(FN) were extracted for analysis. Finally, 5 articles were selected for our meta-analysis. STATA version 15 was used to analysis data. Result: Our meta-analysis showed a pooled sensitivity of 89%(95% CI: 81%-94%)with low heterogeneity(I2 = 17.51%), and a pooled specificity of 89% (95% CI: 71%-96%), with significant heterogeneity(I2 = 95.87%). The pooled positive diagnostic likelihood ratio(pDLR) was 7.81(95% CI: 2.82-21.61, I2 = 92.66%), and the negative diagnostic likelihood ratio(nDLR) was 0.12(95% CI: 0.07-0.22, I2 = 11.94%). Conclusion: Diagnostic MRI characteristics such as NAC enhancement, non-mass enhancement(NME) type, mass size (greater than 20 mm), and tumor-to-nipple distance(TND) were assocaited with the presence of NAC involement. Because of high diagnostic accuracy, MRI imaging can be used as a helpful tool for NAC diagnosis.

In computer-aided detection of microcalcifications (MCs), the detection accuracy is often compromised by frequent occurrence of false positives (FPs), which can be attributed to a number of factors, including imaging noise, inhomogeneity in tissue background, linear structures, and artifacts in mammograms. In this study, the authors investigated a unified classification approach for combating the adverse effects of these heterogeneous factors for accurate MC detection. To accommodate FPs caused by different factors in a mammogram image, the authors developed a classification model to which the input features were adapted according to the image context at a detection location. For this purpose, the input features were defined in two groups, of which one group was derived from the image intensity pattern in a local neighborhood of a detection location, and the other group was used to characterize how a MC is different from its structural background. Owing to the distinctive effect of linear structures in the detector response, the authors introduced a dummy variable into the unified classifier model, which allowed the input features to be adapted according to the image context at a detection location (i.e., presence or absence of linear structures). To suppress the effect of inhomogeneity in tissue background, the input features were extracted from different domains aimed for enhancing MCs in a mammogram image. To demonstrate the flexibility of the proposed approach, the authors implemented the unified classifier model by two widely used machine learning algorithms, namely, a support vector machine (SVM) classifier and an Adaboost classifier. In the experiment, the proposed approach was tested for two representative MC detectors in the literature [difference-of-Gaussians (DoG) detector and SVM detector]. The detection performance was assessed using free-response receiver operating characteristic (FROC) analysis on a set of 141 screen-film mammogram (SFM) images (66 cases) and a set of 188 full-field digital mammogram (FFDM) images (95 cases). The FROC analysis results show that the proposed unified classification approach can significantly improve the detection accuracy of two MC detectors on both SFM and FFDM images. Despite the difference in performance between the two detectors, the unified classifiers can reduce their FP rate to a similar level in the output of the two detectors. In particular, with true-positive rate at 85%, the FP rate on SFM images for the DoG detector was reduced from 1.16 to 0.33 clusters/image (unified SVM) and 0.36 clusters/image (unified Adaboost), respectively; similarly, for the SVM detector, the FP rate was reduced from 0.45 clusters/image to 0.30 clusters/image (unified SVM) and 0.25 clusters/image (unified Adaboost), respectively. Similar FP reduction results were also achieved on FFDM images for the two MC detectors. The proposed unified classification approach can be effective for discriminating MCs from FPs caused by different factors (such as MC-like noise patterns and linear structures) in MC detection. The framework is general and can be applicable for further improving the detection accuracy of existing MC detectors.

Autism spectrum disorder (ASD) is characterized by structural and functional brain changes that contribute to memory, attention and social interaction. The aim of this research is to develop a deep learning framework using Siamese neural nets for computer aided diagnosis of ASD using T1-weighted magnetic resonance imaging (MRI) of 102 control and 112 ASD patients from autism brain imaging data exchange. The preprocessing of the images involves reorientation to a standard space, cropping followed by affine registration to a template. Siamese Neural Network (SNN) with pre-trained ResNet50 model was employed for this study. After preprocessing, the affine registered images are down sampled and reshaped to match with the required input size of the ResNet50. Further, 1070 positive and negative image pairs are formed for training and validation of the SNN model. Final layer of ResNet50 is global averaged and an extra dense layer is added which represents the input image embedding. Further, L1-distance is computed between the embeddings of the two inputs which is further used to backpropagate the error computed using the contrastive loss function. The quality metrics used during 5-fold stratified cross-validation are accuracy, recall, precision and f1-score and these metrics reached a value of 0.99 during validation. Therefore, the developed SNN based tool could be used for diagnosis of autism from T1-weighted MRI.

This study investigates the value of magnetic resonance imaging (MRI) based on a deep learning algorithm in the diagnosis of diabetic macular edema (DME) patients. A total of 96 patients with DME were randomly divided into the experimental group (N = 48) and the control group (N = 48). A deep learning 3D convolutional neural network (3D-CNN) algorithm for MRI images of patients with DME was designed. The application value of this algorithm was comprehensively evaluated by MRI image segmentation Dice value, sensitivity, specificity, and other indicators and diagnostic accuracy. The results showed that the quality of MRI images processed by the 3D-CNN algorithm based on deep learning was significantly improved, and the Dice value, sensitivity, and specificity index data were significantly better than those of the traditional CNN algorithm (P < 0.05). In addition, the diagnostic accuracy of MRI images processed by this algorithm was 93.78 ± 5.32%, which was significantly better than the diagnostic accuracy of 64.25 ± 10.24% of traditional MRI images in the control group (P < 0.05). In summary, the 3D-CNN algorithm based on deep learning can significantly improve the accuracy and sensitivity of MRI image recognition and segmentation in patients with DME, can significantly improve the diagnostic accuracy of MRI in patients with DME, and has a good clinical application value.

Anonymous communication serves as the preferred tool for cyber attackers to evade detection, posing a serious threat to cyberspace security. Accurately tracking the attackers in anonymous communication is crucial for defending against attacks. Flow correlation is an effective method that can link flows in the anonymous network. Existing flow correlation methods usually rely on a long observation, resulting in reduced correlation precision and limited generalization ability within anonymous communication. To address this issue, we propose a robust two-stage flow correlation method called RTsFCM via Siamese network and ensemble voting scheme. In the first stage, a Siamese network with shared weights is utilized to automatically extract the multilevel features from ingress flow and egress flow, respectively. Further, they are concatenated to generate a more expressive feature set to enhance true positive rate (TPR). In the second stage, flow pairs are firstly divided into a series of partially overlapping sub-flows(windows) in view of flow duration. Then, pairwise comparison for each window is conducted independently and the ensemble voting scheme is adopted across these windows to reduce the false positive rate (FPR) significantly. Experimental results show that RTsFCM is superior to the state of the art, achieving over a 4% increase in both TPR and F1_score. Simultaneously, it obtains an FPR as low as 0.68%, utilizing the packet timing characteristics within the initial portion of a flow.

Artificial neural networks have been applied to the differentiation of masses from false- positive detections in digital mammograms. A database of 110 pairs of digital mammograms containing a total of 102 masses (54 malignant, 48 benign) was utilized in this study. Three- hundred-two false positive regions were selected from these images to be used in the training of the artificial neural network. Over 90 features were calculated for both the true masses and the false positives. Features that showed the most separation, in a one-dimensional analysis, between true positives and false positives were selected for artificial neural network input. A three level feed-forward neural network was used with one input layer, one hidden layer and an output layer. By varying the structure and learning rate of the neural network an optimal structure was found. The performance of the ANN was evaluated by means of receiver operating characteristic (ROC) analysis and free-response receiver operating characteristic (FROC) analysis. Results from a round robin evaluation yielded an Az of 0.97 in the task of differentiating between masses and false-positive detections. In the future, multi-dimensional feature analysis will be performed to obtain the optimal performance using a combination of rule-based decision making along with artificial neural networks.

Possible MRI-Mastectomy Link Sparks Debate on MRI's Role in Breast Cancer Management

Introduction. Detection of liver metastases is important to guide therapeutic management of pancreatic cancer. Data suggest that magnetic resonance imaging (MRI) is more sensitive than computed tomography (CT) for the diagnosis of liver metastases from pancreatic cancer. However, MRI is currently not recommended in several guidelines. Purpose. To compare the performance of CT vs MRI in detection of liver metastases in patients with pancreatic adenocarcinoma. Methods. This was a retrospective single-center study over 24 months from January 2020 to December 2021 of patients suspected of having PDAC. All patients underwent contrast-enhanced triple phase CT and contrast-enhanced MRI/MRCP. The CT and MRI assessment of resectability and detection of metastases were performed according to NCCN version 1.2024 criteria by one observer based on clinical report. Composite reference standard based on decision of operability (based on multidisciplinary tumor board decision), surgical findings (delay of 2 to 3 weeks after imaging) and histopathologic findings. Results. 117 patients (M/F 52/65, mean age 62±11y) were included. The mean tumor size was 43.6±15.2mm (add range). 42/46 patients (34.2%/39.3%) were diagnosed with metastatic disease on CT/MRI, 40/36 patients (34.2%/30.8%) were resectable, 20/16 patients (17.1%/13.7%) were borderline resectable and 15/19 patients (12.8%/16.2%) were locally advanced. The detection of liver metastases on CT and MRI was compared with surgical findings in 45 patients (38.5%). The analysis showed that MRI had superior performance to CT, with sensitivity at 100% vs 60% and specificity at 97.1% vs 97.1%, PPV at 90.9% vs 85.7% and NPV 100% vs 89.5% (MRI/CT). Conclusion. MRI is more sensitive than CT in detection of liver metastases in patients with pancreatic adenocarcinoma.

Effect of CAD on Radiologists' Detection of Lung Nodules on Thoracic CT Scans: Analysis of an Observer Performance Study by Nodule Size

Accurately separating healthy tissue from tumorous regions is crucial for effective diagnosis and treatment planning based on magnetic resonance imaging (MRI) data. Current manual detection methods rely heavily on human expertise, so MRI-based segmentation is essential to improving diagnostic accuracy and treatment outcomes. The purpose of this paper is to compare the performance of detecting brain tumors from MRI images through segmentation using an unmodified and modified U-Net architecture from deep neural network (DNN) that has been modified by adding batch normalization and dropout on the encoder layer with and without the freeze layer. The study utilizes a public 2D brain tumor dataset containing 3064 T1-weighted contrast-enhanced images of meningioma, glioma, and pituitary tumors. Model performance was evaluated using intersection over union (IoU) and standard metrics such as precision, recall, f1-score, and accuracy across training, validation, and testing stages. Statistical analysis, including ANOVA and Duncan's multiple range test, was conducted to determine the significance of performance differences across the architectures. Results indicate that while the modified architectures show improved stability and convergence, the freeze layer model demonstrated superior IoU and efficiency, making it a promising approach for more accurate and efficient brain tumor segmentation. The comparison of the three methods revealed that the modified U-Net architecture with a freeze layer significantly reduced training time by 81.72% compared to the unmodified U-Net while maintaining similar performance across validation and testing stages. All three methods showed comparable accuracy and consistency, with no significant differences in performance during validation and testing.

To ensure the continuous supply of power in remote areas utilizing renewable energy resources is significant. Hence, in this research, an effective energy management method for a small-scale hybrid wind-solar-battery-Ultra-capacitor-based microgrid is proposed. Hybrid Energy Storage System (HESS) has been presented in conjunction with wind and solar energy conversion technologies characterized by coupling of power electronic converters, neural networks, optimization, battery, controllers, and ultra-capacitor storage systems to attain the intended performance. The power balance is regulated via an energy management system by considering the fluctuation in load demand and renewable energy power generation. Further, the voltage controller utilizes the deep Siamese neural network (deep SNN) that effectively carries out energy management among the hybrid renewable energy sources. The smelling-based hunting optimization assists in optimal parameter tuning and training of the deep SNN for enhancing the HESS’s efficiency. In addition, the microgrid operates independently and offers a testing area for different energy management systems and testing scenarios. The proposed small-scale microgrid, which is based on renewable energy, can serve as a significant testing area for methods utilized in smart grid applications. The proposed SBHO model’s efficiency is determined by varying the voltage, current, and power of the wind, solar, battery, and ultra-capacitor measurements. The current capacity of the battery reaches −11.06A and the battery voltage reaches 259.831 V in 0.82 s. The DC load measurement utilizing the SBHO approach obtained a DC bus voltage of 357.11V, a load current of 3.348 A within 0.82 s, and in DC power load attained the 1195.58 W within 0.82 s. The battery’s SOC by applying the smelling-based hunting optimization is gradually increased to 50.008% in 0.82s.In terms of the PV measurement, the PV current, PV voltage, and PV power are obtained as 4.238A,241.08V, and 1021.64W for the SBHO approach which surpasses other competent techniques. The ultra-capacitor current ranges from 35A to 40A with reduced heavy discharge of SOC from 98% obtained on evaluating the performance. The output power of wind using a boost convert remains 3000 W with few harmonics.

In recent times, deep learning methods have been employed to learn anatomical and functional brain changes from high discriminative features extracted from neuroimaging data such as Magnetic Resonance Imaging (MRI) which can enhance the performance in the classification and early diagnosis of neurodegenerative diseases. However, features that exist between brain regions that are farther apart are usually not captured by most state-of-the-art deep learning methods. Thus, an effective and robust model for the extraction of high-dimensional descriptive features especially from brain MRI remains an open challenge. In this paper, we investigate the applicability of an enhanced 3D Residual Network (ResNet) for the extraction of high-dimensional descriptive features for an improved classification of neurodegenerative disease using MRI scans. In particular, we enhanced the ResNet-18 by using a dilated convolutional layer instead of the typical convolution layer to expand the receptive field for effective feature extraction and an attention mechanism to the residual blocks to help focus on the relevant extracted features for improved classification. Our proposed method was evaluated on MRI images from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Three MRI scan groups were considered: Alzheimer's Disease (AD), Mild Cognitive Impairment (MCI), and Normal Cognitive (NC). Meanwhile, a three-binary classification task was developed (AD vs. NC, AD vs. MCI, and NC vs. MCI) to test the efficacy of our proposed model. The accuracy of our proposed model for each binary task is 92.12%, 74.07%, and 87.16%, respectively. We further compared the robustness of our proposed model to two state-of-the-art architectures and our model performed better due to its ability to extract discriminative features from the MRI data relevant for the classification tasks. Thus, revealing the effectiveness of our proposed method on the MRI scans.