An Improved Fully Automated Breast Cancer Detection and Classification System

More than 500,000 patients are diagnosed with breast cancer annually. Authorities worldwide reported a death rate of 11.6% in 2018. Breast tumors are considered a fatal disease and primarily affect middle-aged women. Various approaches to identify and classify the disease using different technologies, such as deep learning and image segmentation, have been developed. Some of these methods reach 99% accuracy. However, boosting accuracy remains highly important as patients’ lives depend on early diagnosis and specified treatment plans. This paper presents a fully computerized method to detect and categorize tumor masses in the breast using two deep-learning models and a classifier on different datasets. This method specifically uses ResNet50 and AlexNet, convolutional neural networks (CNNs), for deep learning and a K-Nearest-Neighbor (KNN) algorithm to classify data. Various experiments have been conducted on five datasets: the Mammographic Image Analysis Society (MIAS), Breast Cancer Histopathological Annotation and Diagnosis (BreCaHAD), King Abdulaziz University Breast Cancer Mammogram Dataset (KAU-BCMD), Breast Histopathology Images (BHI), and Breast Cancer Histopathological Image Classification (BreakHis). These datasets were used to train, validate, and test the presented method. The obtained results achieved an average of 99.38% accuracy, surpassing other models. Essential performance quantities, including precision, recall, specificity, and F-score, reached 99.71%, 99.46%, 98.08%, and 99.67%, respectively. These outcomes indicate that the presented method offers essential aid to pathologists diagnosing breast cancer. This study suggests using the implemented algorithm to support physicians in analyzing breast cancer correctly.     

改进注意力机制的滚动轴承故障诊断方法研究

针对滚动轴承在实际工作环境中噪声较大和负载变化的问题,提出一种基于双注意卷积机制的残差神经网络(Double attention convolution mechanism ResNet,DACM_ResNet)轴承故障诊断方法：首先,对滚动轴承振动信号进行短时傅里叶变换(short-time Fourier transform,STFT)并使用伪彩色处理得到三通道图像数据；然后,对残差神经网络在轴承故障诊断上进行研究,在残差单元的卷积层之后,使用DACM模块,将残差特征在通道和空间维度上进行进一步提取,最后,在凯斯西储大学(CWRU)数据集上进行试验验证,试验结果表明所提出的方法在噪声环境下及负载变化时,平均诊断准确率达到了98%以上,说明所提出的模型有较好的鲁棒性。     

Modelling a Fused Deep Network Model for Pneumonia Prediction

Deep Learning (DL) is known for its golden standard computing paradigm in the learning community. However, it turns out to be an extensively utilized computing approach in the ML field. Therefore, attaining superior outcomes over cognitive tasks based on human performance. The primary benefit of DL is its competency in learning massive data. The DL-based technologies have grown faster and are widely adopted to handle the conventional approaches resourcefully. Specifically, various DL approaches outperform the conventional ML approaches in real-time applications. Indeed, various research works are reviewed to understand the significance of the individual DL models and some computational complexity is observed. This may be due to the broader expertise and knowledge required for handling these models during the prediction process. This research proposes a holistic approach for pneumonia prediction and offers a more appropriate DL model for classification purposes. This work incorporates a novel fused Squeeze and Excitation (SE) block with the ResNet model for pneumonia prediction and better accuracy. The expected model reduces the human effort during the prediction process and makes it easier to diagnose it intelligently as the feature learning is adaptive. The experimentation is carried out in Keras, and the model’s superiority is compared with various advanced approaches. The proposed model gives 90% prediction accuracy, 93% precision, 90% recall and 89% F1-measure. The proposed model shows a better trade-off compared to other approaches. The evaluation is done with the existing standard ResNet model, GoogleNet+ResNet+DenseNet, and different variants of ResNet models.     

基于优化 CBAM 改进 ResNet50 的异常 行为识别方法

在自动视频监控应用中,准确地识别出人类的异常行为是非常困难的任务。为了解决监测系统中异常人类活动的高 效识别问题,提出了一种加强局部以及全局特征信息融合的异常行为识别模型 ICBAM-ResNet50 。在 UTI 和 CASIA 两个数 据集上进行实验,结果表明该研究比ResNet50 模型准确率分别提高了7%和8%。ICBAM 模块引入一维卷积替换了原始 CBAM 中通道注意力的 MLP 操作,将局部的时间特征整合到通道描述符中,缓解了通道维度由于全局处理产生的忽略信息 交互的问题;其次引入时空注意力机制替换 CBAM 中的单一空间注意力机制,来提高模型的时空表征能力。最后,将优化的 CBAM 模块嵌入到 ResNet50 中,通过在 ImageNet 上对其进行预训练,在两个基准数据集上该模型分别达到了98.8%和97.9%的准确率。使用相同的数据集,将实验结果与原始识别方法进行了比较,结果表明该模型优于所比较的其他方法。     

深度学习在分组密码差分区分器上的研究应用

差分分析在分组密码分析领域是一种重要的研究方法, 针对分组密码的差分分析的重点在于找到一个轮数或者概率更大的差分区分器. 首先描述了通过深度学习技术构造差分区分器时所需要的数据集的构造方法, 并且分别基于卷积神经网络(convolutional neural networks, CNN)和残差神经网络(residual...