To address the issue in the existing Whole Slide Image （WSI） classification for lung cancer， where Multi-Instance Learning （MIL） methods based on knowledge distillation architecture struggle to handle the mixture of hard and easy instances effectively， leading to insufficient model learning capability and imbalanced performance， a lung cancer whole slide image classification method based on Bidirectional Collaborative Distillation and Multi-Instance Learning （BCD-MIL） was proposed by incorporating the concept of multi-task learning. First， a Bidirectional Collaborative Distillation Framework （BCDF） comprising multiple student models and a teacher model was designed， so as to satisfy the differentiated learning requirements for easy and hard pathological instances in lung cancer histopathological images， and by distributing different tasks to multiple student models， the model's learning capacity was enhanced while solving performance imbalance problem. Concurrently， classification performance was improved through a virtuous cycle with the teacher model. Second， a Multi-Angle Instance Mining Parallel architecture （MA-IMP） was designed to match the heterogeneous differences in lung cancer pathological features across dimensions such as cell morphology and tissue texture， and by conducting instance mining from multiple perspectives， the mining bias caused by a single perspective was avoided. Finally， a Dynamic Stage-Aware Exponential Moving Average distillation （DSA-EMA） algorithm was proposed to optimize weight update of the teacher model， and improve training efficiency and model performance based on the stage characteristics of large-scale instance training of lung cancer histopathological images through adjusting distillation parameters in the training phase dynamically. Experimental results show that compared to the MIL framework with Masked Hard Instance Mining （MHIM-MIL） method， which is also based on knowledge distillation and instance mining， BCD-MIL achieves improvements of 1.92， 2.37， 3.43， and 1.75 percentage points in Area Under Curve （AUC）， accuracy， F1-Score， and recall， respectively， on The Cancer Genome Atlas （TCGA） dataset， and improvements of 1.11， 6.29， 6.94， and 12.85 percentage points in four metrics， respectively， on the Clinical Proteomic Tumor Analysis Consortium （CPTAC） dataset； validating the effectiveness of the proposed method. Furthermore， the lightweight distillation architecture reduces model parameter size and inference time， thereby enabling efficient deployment while ensuring performance gains， and providing a reliable basis for lung cancer WSI classification.
Elastic registration is regarded as a key task in medical image processing， whose performance directly affects the accuracy of subsequent tasks such as segmentation， classification， and prediction. However， due to the insensitivity of neural networks to high-frequency components， the existing methods have difficulty in capturing high-frequency information in images， which affects the fitting accuracy of registration field. To address this issue， a high-frequency-preserving medical image registration model based on frequency spectrum decomposition — DFRes （Decomposition in Frequency domain model for Registration） was proposed. In the model， a frequency decomposition strategy was introduced， and a dual-branch structure was adopted to process high- and low-frequency information from the original image. Meanwhile， an Invertible Neural Network （INN） structure with high-frequency preservation characteristics and a bridge-style feature fusion module with ability to fuse high- and low-frequency information were designed， and an alternating spatial-frequency information extraction module was used to further enhance the model’s ability to extract and fuse frequency- and spatial-domain information. Experimental results of comparing DFRes and the existing advanced models on the IXI， OSSAI， and Huaxi rectal cancer datasets show that DFRes achieves significant improvements on multiple metrics. On IXI dataset， compared to the TransMorph model， DFRes has the Dice Similarity Coefficient （DSC） increased by 2.5 percentage points， the Average Surface Distance （ASD） reduced by 0.012， and the Structural SIMilarity （SSIM） increased by 1.6 percentage points. At the same time， the effectiveness of the module design is verified through ablation experiments.
As a system that converts IP addresses and domain names to each other， Domain Name System （DNS） is one of the important basic protocols in Internet. Due to the importance of DNS in Internet， the security policies of some security facilities such as firewalls and Intrusion Detection Systems （IDSs） allow DNS traffic to pass by default， giving attackers the opportunity to use DNS tunneling for communication. Currently， there are many malware that support DNS communication or even use DNS communication by default， which brings great challenges to network security tools and security operations centers. However， the existing research mainly focuses on specific detection methods and rarely explores the tunneling tools themselves， even though the majority of researchers rely on tunneling tools to generate samples. Therefore， the research on DNS tunnel detection technology was reviewed. Firstly， the development history and research status and the existing detection schemes of DNS tunneling were elaborated systematically， and the advantages and disadvantages of detection methods in the past 10 years were discussed. Subsequently， 6 commonly used tools in these detection schemes such as dnscat2， Iodine， and dns2tcp were evaluated and tested， and the experimental data was published. Experimental results show that most detection schemes do not disclose their tunneling sample datasets or the set parameters when using tunneling tools to generate traffic， making these schemes almost impossible to reproduce. Besides， some detection solutions use DNS tunneling tools with distinctive signature characteristics. Using samples with signature features to train model-based detection schemes will lead to doubts about the generalization ability of the model， that is， it is impossible to know whether this type of model will perform well in the real world. Finally， related future work development directions were prospected.
Differentiable ARchiTecture Search （DARTS） can design neural network architectures efficiently and automatically. However， there is a performance “wide gap” between the construction method of super network and the design of derivation strategy in it. To solve the above problem， a differentiable neural architecture search algorithm with constraint in optimal search space was proposed. Firstly， the training process of the super network was analyzed by using the architecture parameters associated with the candidate operations as the quantitative indicators， and it was found that the invalid candidate operation none occupied the architecture parameter with the maximum weight in deviation architecture， which caused that architectures obtained by the algorithm had poor performance. Aiming at this problem， an optimized search space was proposed. Then， the difference between the super network of DARTS and derivation architecture was analyzed， the architecture entropy was defined based on architecture parameters， and this architecture entropy was used as the constraint of the objective function of DARTS， so as to promote the super network to narrow the difference with the derivation strategy. Finally， experiments were conducted on CIFAR-10 dataset. The experimental results show that the searched architecture by the proposed algorithm achieved 97.17% classification accuracy in these experiments， better than the comparison algorithms in accuracy， parameter quantity and search time comprehensively. The proposed algorithm is effective and improves classification accuracy of searched architecture on CIFAR-10 dataset.
All market activities of stock market participants combine to affect stock market changes， making stock market volatility fraught with complexity and making accurate prediction of stock prices a challenge. Among these activities that affect stock market changes， financial disclosure is an attractive and potentially financially rewarding means of predicting stock indexe changes. In order to deal with the complex changes in the stock market， a method of stock index prediction was proposed that incorporates data from financial statements disclosed by corporates. Firstly， the stock index historical data and corporate financial statement data were preprocessed， and the main task is dimension reduction of the high-dimensional matrix generated from corporate financial statement data， and then the dual-channel Long Short-Term Memory （LSTM） network was used to forecast and research the normalized data. Experimental results on SSE 50 and CSI 300 Index datasets show that the prediction effect of the proposed method is better than that using only historical data of stock indexes.
To solve the problem that high dimension of descriptor decreases the matching speed of Scale Invariant Feature Transform (SIFT) algorithm, an improved SIFT algorithm was proposed. The feature point was acted as the center, the circular rotation invariance structure was used to construct feature descriptor in the approximate size circular feature points' neighborhood, which was divided into several sub-rings. In each sub-ring, the pixel information was to maintain a relatively constant and positions changed only. The accumulated value of the gradient within each ring element was sorted to generate the feature vector descriptor when the image was rotated. The dimensions and complexity of the algorithm was reduced and the dimensions of feature descriptor were reduced from 128 to 48. The experimental results show that, the improved algorithm can improve rotating registration repetition rate to more than 85%. Compared with the SIFT algorithm, the average matching registration rate increases by 5%, the average time of image registration reduces by about 30% in the image rotation, zoom and illumination change cases. The improved SIFT algorithm is effective.
