Search

Background: Gene status has become the focus of prognosis prediction. Furthermore, deep learning has frequently been implemented in medical imaging to diagnose, prognosticate, and evaluate treatment responses in patients with cancer. However, few deep learning survival (DLS) models based on mutational genes that are directly associated with patient prognosis in terms of progression-free survival (PFS) or overall survival (OS) have been reported. Additionally, DLS models have not been applied to determine IO-related prognosis based on mutational genes. Herein, we developed a deep learning method to predict the prognosis of patients with lung cancer treated with or without immunotherapy (IO). Methods: Samples from 6542 patients from different centers were subjected to genome sequencing. A DLS model based on multi-panels of somatic mutations was trained and validated to predict OS in patients treated without IO and PFS in patients treated with IO. Results: In patients treated without IO, the DLS model (low vs. high DLS) was trained using the training MSK-MET cohort (HR = 0.241 [0.213–0.273], P < 0.001) and tested in the inter-validation MSK-MET cohort (HR = 0.175 [0.148–0.206], P < 0.001). The DLS model was then validated with the OncoSG, MSK-CSC, and TCGA-LUAD cohorts (HR = 0.420 [0.272–0.649], P < 0.001; HR = 0.550 [0.424–0.714], P < 0.001; HR = 0.215 [0.159–0.291], P < 0.001, respectively). Subsequently, it was fine-tuned and retrained in patients treated with IO. The DLS model (low vs. high DLS) could predict PFS and OS in the MIND, MSKCC, and POPLAR/OAK cohorts (P < 0.001, respectively). Compared with tumor-node-metastasis staging, the COX model, tumor mutational burden, and programmed death-ligand 1 expression, the DLS model had the highest C-index in patients treated with or without IO. Conclusions: The DLS model based on mutational genes can robustly predict the prognosis of patients with lung cancer treated with or without IO. [ABSTRACT FROM AUTHOR]

With the wide application of remote sensing technology, target detection based on deep learning has become a research hotspot in the field of remote sensing. In this paper, aimed at the problems of the existing deep-learning-based desert land intelligent extraction methods, such as the spectral similarity of features and unclear texture features, we propose a multispectral remote sensing image desert land intelligent extraction method that takes into account band information. Firstly, we built a desert land intelligent interpretation dataset based on band weighting to enhance the desert land foreground features of the images. On this basis, we introduced the deformable convolution adaptive feature extraction capability to U-Net and developed the Y-Net model to extract desert land from Landsat remote sensing images covering the Inner Mongolia Autonomous Region. Finally, in order to analyze the spatial and temporal trends of the desert land in the study area, we used a structural equation model (SEM) to evaluate the direct and indirect effects of natural conditions and human activities, i.e., population density (PD), livestock volume (LS), evaporation (Evp), temperature (T), days of sandy wind conditions (LD), humidity (RH), precipitation (P), anthropogenic disturbance index (Adi), and cultivated land (CL). The results show that the F1-score of the Y-Net model proposed in this paper is 95.6%, which is 11.5% more than that of U-Net. Based on the Landsat satellite images, the area of desert land in the study area for six periods from 1990 to 2020 was extracted. The results show that the area of desert land in the study area first increased and then decreased. The main influencing factors have been precipitation, humidity, and anthropogenic disturbance, for which the path coefficients are 0.646, 0.615, and 0.367, respectively. This study will be of great significance in obtaining large-scale and long-term time series of desert land cover and revealing the inner mechanism of desert land area change. [ABSTRACT FROM AUTHOR]

In this study, a general network model called multi-progressive image deblurring network is proposed to correct blurring artifacts and local imaging details in underwater images. As a solution to nonuniform image distortion, a deformable convolution module was designed to enrich the encoded information of the image representation. Using a stepwise feature refinement module, multi-progressive image deblurring network can reduce the loss of contextual information to produce a more realistic underwater image for subsequent applications. Constructing a loss function based on multi-scale content can help the model improve image perception quality. We conducted experimental evaluations on large-scale image deblurring benchmark datasets, such as GoPro and HIDE, achieving excellent results with 32.84 dB and 31.03 dB peak signal-to-noise ratio, respectively, using the proposed method. Subsequently, a detailed optimization comparison was conducted on the in-house underwater image deblurring dataset. Multi-progressive image deblurring network obtained higher-quality, clearer images. Compared with the current state-of-the-art image deblurring algorithms, the proposed model achieved significant results with a 6.6% increase in deblur performance in peak signal-to-noise ratio. Finally, we conducted ablation experiments to evaluate the effectiveness of all the modules in the proposed framework. • Multi-progressive structure can hierarchically fuse global context information. • MPDNet outperforms almost all the SOTA methods on the benchmark datasets. • MPDNet is enough lightweight for deblurring tasks on mobile computing platforms. [ABSTRACT FROM AUTHOR]

The lack of large-scale, multi-scene, and multi-type pavement distress training data reduces the generalization ability of deep learning models in complex scenes, and limits the development of pavement distress extraction algorithms. Thus, we built the first large-scale dichotomous image segmentation (DIS) dataset for multi-type pavement distress segmentation, called ISTD-PDS7, aimed to segment highly accurate pavement distress types from natural charge-coupled device (CCD) images. The new dataset covers seven types of pavement distress in nine types of scenarios, along with negative samples with texture similarity noise. The final dataset contains 18,527 images, which is many more than the previously released benchmarks. All the images are annotated with fine-grained labels. In addition, we conducted a large benchmark test, evaluating seven state-of-the-art segmentation models, providing a detailed discussion of the factors that influence segmentation performance, and making cross-dataset evaluations for the best-performing model. Finally, we investigated the effectiveness of negative samples in reducing false positive prediction in complex scenes and developed two potential data augmentation methods for improving the segmentation accuracy. We hope that these efforts will create promising developments for both academics and the industry. [ABSTRACT FROM AUTHOR]

It is critical to detect malicious code for the security of the Internet of Things (IoT). Therefore, this work proposes a malicious code detection algorithm based on the novel feature fusion–malware image convolutional neural network (FF-MICNN). This method combines a feature fusion algorithm with deep learning. First, the malicious code is transformed into grayscale image features by image technology, after which the opcode sequence features of the malicious code are extracted by the n-gram technique, and the global and local features are fused by feature fusion technology. The fused features are input into FF-MICNN for training, and an appropriate classifier is selected for detection. The results of experiments show that the proposed algorithm exhibits improvements in its detection speed, the comprehensiveness of features, and accuracy as compared with other algorithms. The accuracy rate of the proposed algorithm is also 0.2% better than that of a detection algorithm based on a single feature. [ABSTRACT FROM AUTHOR]