基于深度残差神经网络迁移学习的牙形刺图像识别

为了探索基于深度神经网络模型的牙形刺图像智能识别效果,研究选取奥陶纪8种牙形刺作为研究对象,通过体视显微镜采集牙形刺图像1188幅,收集整理公开发表文献的牙形刺图像778幅,将图像数据集划分为训练集和测试集。通过对训练集图像进行旋转、翻转、滤波增强处理,解决了训练样本不足的问题。基于ResNet-18、ResNet-34、ResNet-50、ResNet-101、ResNet-152五种残差神经网络模型,采用迁移学习方法,对网络模型进行训练以获取模型参数,五种模型测试Top-1准确率分别为85.37%、85.85%、83.90%、81.95%、80.00%, Top-2准确率分别为94.63%、94.63%、94.15%、93.17%、93.66%,模型对牙形刺图像具有较好的识别效果。通过对比研究发现,ResNet-34识别准确率最高,说明对于特征简单的牙形刺属种,增加网络深度并不一定能提升准确率,而确定合适深度的模型则不仅可以提高识别准确率,还可以节约计算资源。通过ResNet-34模型的迁移学习训练和重新训练效果对比可以看出,迁移学习不仅可以获得较高的准确率,而且可以较快获取模型参...     

Visual and buying sequence features-based product image recommendation using optimization based deep residual network

A recommendation system is an imaginative resolution for managing the restrictions in e-commerce services with item details and user details. Also, it is used to determine the user preferences to recommend the items they expected to buy. Several conventional collaborative filtering techniques are devised in the recommender model, but it has some complexities. Hence, an innovative optimization-driven deep residual network is devised in this paper for a product recommendation system. Here, the product of images is used for extracting features where the Convolutional neural network (CNN) features are computed, and then it is given as input to the deep residual network aimed at product recommendation. The deep residual network is trained using developed Elephant Herding Feedback Artificial Optimization (EHFAO), which is obtained by integrating Elephant Herding optimization (EHO) into the Feedback Artificial Tree (FAT). Here, the item grouping is carried out on input data based on K-means clustering. After item grouping, Cosine similarity is used to perform matching of groups, where the best group is acquired among all the available groups. Extraction of list of visitors is done from the best group. Then, the list of items is obtained from the sequence of best visitor. Next, the corresponding binary sequence is obtained for the applicable sequence of visitor. From this sequence of best visitor, the recommended product is acquired. Then, the recommended product is subjected to the sentiment analysis for which the score is determined. Here, the sentiment analysis helps to decide whether the product is recommended or not recommended. If the score is positive, then the same product is recommended; otherwise, the new product is recommended. The proposed EHFAO-based deep residual network attained better performance in comparison to the other techniques with a maximal F-measure at 84.061%, 84.061% precision, 87.845% recall along with minimal Mean Squared Error (MSE) of 0.216.     

地高辛标记探针检测生物制品中残余DNA含量

用地高辛标记探针检测由传代细胞系生产的人用精制狂犬病疫苗,重组（CHO细胞）乙肝疫苗,出血热疫苗及痢疾多糖结合疫苗原液中残余DNA含量。结果表明,该方法特异性强,灵敏度高,可用于上述生物制品中残余DNA含量的检测。     

T1-weighted and T2-weighted MRI image synthesis with convolutional generative adversarial networks

BackgroundThe objective of this study was to propose an optimal input image quality for a conditional generative adversarial network (GAN) in T1-weighted and T2-weighted magnetic resonance imaging (MRI) images.Materials and methodsA total of 2,024 images scanned from 2017 to 2018 in 104 patients were used. The prediction framework of T1-weighted to T2-weighted MRI images and T2-weighted to T1-weighted MRI images were created with GAN. Two image sizes (512 × 512 and 256 × 256) and two grayscale level conversion method (simple and adaptive) were used for the input images. The images were converted from 16-bit to 8-bit by dividing with 256 levels in a simple conversion method. For the adaptive conversion method, the unused levels were eliminated in 16-bit images, which were converted to 8-bit images by dividing with the value obtained after dividing the maximum pixel value with 256.ResultsThe relative mean absolute error (rMAE ) was 0.15 for T1-weighted to T2-weighted MRI images and 0.17 for T2-weighted to T1-weighted MRI images with an adaptive conversion method, which was the smallest. Moreover, the adaptive conversion method has a smallest mean square error (rMSE) and root mean square error (rRMSE), and the largest peak signal-to-noise ratio (PSNR) and mutual information (MI). The computation time depended on the image size.ConclusionsInput resolution and image size affect the accuracy of prediction. The proposed model and approach of prediction framework can help improve the versatility and quality of multi-contrast MRI tests without the need for prolonged examinations.     

基于三维卷积神经网络的肺结节探测与定位方法

提出一种基于三维卷积神经网络对肺部计算机断层扫描图像(CT)进行肺结节自动探测及定位的方法.基于开源数据集LUNA16开展研究,对数据进行像素归一化、坐标转换等预处理,对正样本使用随机平移、旋转和翻转的方式进行扩充,对负样本进行随机采样.搭建了三维卷积神经网络并在训练过程中调整网络参数,直到得到性能最佳的网络.此外还设...     

Optical coherence tomography image denoising using a generative adversarial network with speckle modulation

Optical coherence tomography (OCT) is widely used for biomedical imaging and clinical diagnosis. However, speckle noise is a key factor affecting OCT image quality. Here, we developed a custom generative adversarial network (GAN) to denoise OCT images. A speckle‐modulating OCT (SM‐OCT) was built to generate low speckle images to be used as the ground truth. In total, 210 000 SM‐OCT images were used for training and validating the neural network model, which we call SM‐GAN. The performance of the SM‐GAN method was further demonstrated using online benchmark retinal images, 3D OCT images acquired from human fingers and OCT videos of a beating fruit fly heart. The denoise performance of the SM‐GAN model was compared to traditional OCT denoising methods and other state‐of‐the‐art deep learning based denoise networks. We conclude that the SM‐GAN model presented here can effectively reduce speckle noise in OCT images and videos while maintaining spatial and temporal resolutions.     

Calculating the target exposure index using a deep convolutional neural network and a rule base

PurposeThe objective of this study is to determine the quality of chest X-ray images using a deep convolutional neural network (DCNN) and a rule base without performing any visual assessment. A method is proposed for determining the minimum diagnosable exposure index (EI) and the target exposure index (EIt).MethodsThe proposed method involves transfer learning to assess the lung fields, mediastinum, and spine using GoogLeNet, which is a type of DCNN that has been trained using conventional images. Three detectors were created, and the image quality of local regions was rated. Subsequently, the results were used to determine the overall quality of chest X-ray images using a rule-based technique that was in turn based on expert assessment. The minimum EI required for diagnosis was calculated based on the distribution of the EI values, which were classified as either suitable or non-suitable and then used to ascertain the EIt.ResultsThe accuracy rate using the DCNN and the rule base was 81%. The minimum EI required for diagnosis was 230, and the EIt was 288.ConclusionThe results indicated that the proposed method using the DCNN and the rule base could discriminate different image qualities without any visual assessment; moreover, it could determine both the minimum EI required for diagnosis and the EIt.     

Improvement of image quality for pancreatic cancer using deep learning-generated virtual monochromatic images: Comparison with single-energy computed tomography

PurposeTo construct a deep convolutional neural network that generates virtual monochromatic images (VMIs) from single-energy computed tomography (SECT) images for improved pancreatic cancer imaging quality.Materials and methodsFifty patients with pancreatic cancer underwent a dual-energy CT simulation and VMIs at 77 and 60 keV were reconstructed. A 2D deep densely connected convolutional neural network was modeled to learn the relationship between the VMIs at 77 (input) and 60 keV (ground-truth). Subsequently, VMIs were generated for 20 patients from SECT images using the trained deep learning model.ResultsThe contrast-to-noise ratio was significantly improved (p < 0.001) in the generated VMIs (4.1 ± 1.8) compared to the SECT images (2.8 ± 1.1). The mean overall image quality (4.1 ± 0.6) and tumor enhancement (3.6 ± 0.6) in the generated VMIs assessed on a five-point scale were significantly higher (p < 0.001) than that in the SECT images (3.2 ± 0.4 and 2.8 ± 0.4 for overall image quality and tumor enhancement, respectively).ConclusionsThe quality of the SECT image was significantly improved both objectively and subjectively using the proposed deep learning model for pancreatic tumors in radiotherapy.     

Identification of potential targets in biological signalling systems through network perturbation analysis

The network-based representation and analysis of biological systems contributes to a greater understanding of their structures and functions at different levels of complexity. These techniques can also be used to identify potential novel therapeutic targets based on the characterisation of vulnerable or highly influential network components. There is a need to investigate methods for estimating the impact of molecular perturbations. The prediction of high-impact or critical targets can aid in the identification of novel strategies for controlling the level of activation of specific, therapeutically relevant genes or proteins. Here, we report a new computational strategy for the analysis of the vulnerability of cellular signalling networks based on the quantitative assessment of the impact of large-scale, dynamic perturbations. To show the usefulness of this methodology, two complex signalling networks were analysed: the caspase-3 and the adenosine-regulated calcium signalling systems. This allowed us to estimate and rank the perturbation impact of the components defining these networks. Testable hypotheses about how these targets could modify the dynamic operation of the systems are provided. In the case of the caspase-3 system, the predictions and rankings were in line with results obtained from previous experimental validations of computational predictions generated by a relatively more computationally complex technique. In the case of the adenosine-regulated calcium system, we offer new testable predictions on the potential effect of different targets on the control of calcium flux. Unlike previous methods, the proposed approach provides perturbation-specific scores for each network component. The proposed perturbation assessment methodology may be applied to other systems to gain a deeper understanding of their dynamic operation and to assist the discovery of new therapeutic targets and strategies.     

Multiomics metabolic and epigenetics regulatory network in cancer: A systems biology perspective

Genetic, epigenetic, and metabolic alterations are all hallmarks of cancer. However, the epigenome and metabolome are both highly complex and dynamic biological networks in vivo. The interplay between the epigenome and metabolome contributes to a biological system that is responsive to the tumor microenvironment and possesses a wealth of unknown biomarkers and targets of cancer therapy. From this perspective, we first review the state of high-throughput biological data acquisition(i.e. multiomics data)and analysis(i.e. computational tools) and then propose a conceptual in silico metabolic and epigenetic regulatory network(MER-Net) that is based on these current high-throughput methods. The conceptual MER-Net is aimed at linking metabolomic and epigenomic networks through observation of biological processes, omics data acquisition, analysis of network information, and integration with validated database knowledge. Thus, MER-Net could be used to reveal new potential biomarkers and therapeutic targets using deep learning models to integrate and analyze large multiomics networks. We propose that MER-Net can serve as a tool to guide integrated metabolomics and epigenomics research or can be modified to answer other complex biological and clinical questions using multiomics data.     

A new deep learning method for image deblurring in optical microscopic systems

Deconvolution is the most commonly used image processing method in optical imaging systems to remove the blur caused by the point‐spread function (PSF). While this method has been successful in deblurring, it suffers from several disadvantages, such as slow processing time due to multiple iterations required to deblur and suboptimal in cases where the experimental operator chosen to represent PSF is not optimal. In this paper, we present a deep‐learning‐based deblurring method that is fast and applicable to optical microscopic imaging systems. We tested the robustness of proposed deblurring method on the publicly available data, simulated data and experimental data (including 2D optical microscopic data and 3D photoacoustic microscopic data), which all showed much improved deblurred results compared to deconvolution. We compared our results against several existing deconvolution methods. Our results are better than conventional techniques and do not require multiple iterations or pre‐determined experimental operator. Our method has several advantages including simple operation, short time to compute, good deblur results and wide application in all types of optical microscopic imaging systems. The deep learning approach opens up a new path for deblurring and can be applied in various biomedical imaging fields.     

基于TransformerMGI的microRNA靶向基因预测

生物小分子microRNA可以对基因表达进行正向或负向调控,研究microRNA与基因之间的关系对于机体稳态的维持和疾病治疗都有着重要意义。利用深度学习方法对microRNA和基因靶向关系进行预测,提出了TransformerMGI模型。在特征工程阶段,针对生物序列潜在信息难以准确地提取这一问题,TransformerMGI模型分别采用了基于图卷积神经网络的GP-GCN方法和DNA2Vec模型对microRNA和基因数据的潜在信息进行提取,得到了二者的表征嵌入矩阵,在模型方面,TransformerMGI模型引入了幂归一化来改进经典的深度学习模型。利用microRNA和基因数据经过特征提取后得到两个表征矩阵,这两个矩阵分别被放入TransformerMGI模型中,通过TransformerMGI模型内部的Attention机制对二者自身和相互的特征信息进行了聚合和关联运算,最终预测出microRNA调控基因的概率。采用ROC曲线下面积和准确召回率曲线作为模型性能评价指标,将TransformerMGI与其他现有模型进行了比较评估。实验结果表明,TransformerMGI模型的AUC和AUPRC评分均可达0.91以上,优于现有的其他模型。TransformerMGI模型能在不考虑生物学原理和基因组背景的前提下,仅依赖microRNA和基因的碱基序列信息,实现microRNA靶向基因的预测,从而为后续的microRNA靶向基因预测研究提供了可借鉴的深度学习方法。     

Double-flow convolutional neural network for rapid large field of view Fourier ptychographic reconstruction

Fourier ptychographic microscopy is a promising imaging technique which can circumvent the space-bandwidth product of the system and achieve a reconstruction result with wide field-of-view (FOV), high-resolution and quantitative phase information. However, traditional iterative-based methods typically require multiple times to get convergence, and due to the wave vector deviation in different areas, the millimeter-level full-FOV cannot be well reconstructed once and typically required to be separated into several portions with sufficient overlaps and reconstructed separately, which makes traditional methods suffer from long reconstruction time for a large-FOV (of the order of minutes) and limits the application in real-time large-FOV monitoring of live sample in vitro. Here we propose a novel deep-learning based method called DFNN which can be used in place of traditional iterative-based methods to increase the quality of single large-FOV reconstruction and reducing the processing time from 167.5 to 0.1125 second. In addition, we demonstrate that by training based on the simulation dataset with high-entropy property (Opt. Express 28, 24 152 [2020]), DFNN could has fine generalizability and little dependence on the morphological features of samples. The superior robustness of DFNN against noise is also demonstrated in both simulation and experiment. Furthermore, our model shows more robustness against the wave vector deviation. Therefore, we could achieve better results at the edge areas of a single large-FOV reconstruction. Our method demonstrates a promising way to perform real-time single large-FOV reconstructions and provides further possibilities for real-time large-FOV monitoring of live samples with sub-cellular resolution.     

Novel recurrent neural network for modelling biological networks: Oscillatory p53 interaction dynamics

Understanding the control of cellular networks consisting of gene and protein interactions and their emergent properties is a central activity of Systems Biology research. For this, continuous, discrete, hybrid, and stochastic methods have been proposed. Currently, the most common approach to modelling accurate temporal dynamics of networks is ordinary differential equations (ODE). However, critical limitations of ODE models are difficulty in kinetic parameter estimation and numerical solution of a large number of equations, making them more suited to smaller systems. In this article, we introduce a novel recurrent artificial neural network (RNN) that addresses above limitations and produces a continuous model that easily estimates parameters from data, can handle a large number of molecular interactions and quantifies temporal dynamics and emergent systems properties. This RNN is based on a system of ODEs representing molecular interactions in a signalling network. Each neuron represents concentration change of one molecule represented by an ODE. Weights of the RNN correspond to kinetic parameters in the system and can be adjusted incrementally during network training. The method is applied to the p53-Mdm2 oscillation system – a crucial component of the DNA damage response pathways activated by a damage signal. Simulation results indicate that the proposed RNN can successfully represent the behaviour of the p53-Mdm2 oscillation system and solve the parameter estimation problem with high accuracy. Furthermore, we presented a modified form of the RNN that estimates parameters and captures systems dynamics from sparse data collected over relatively large time steps. We also investigate the robustness of the p53-Mdm2 system using the trained RNN under various levels of parameter perturbation to gain a greater understanding of the control of the p53-Mdm2 system. Its outcomes on robustness are consistent with the current biological knowledge of this system. As more quantitative data become available on individual proteins, the RNN would be able to refine parameter estimation and mapping of temporal dynamics of individual signalling molecules as well as signalling networks as a system. Moreover, RNN can be used to modularise large signalling networks.     

Three critical factors affecting automated image species recognition performance for camera traps

Ecological camera traps are increasingly used by wildlife biologists to unobtrusively monitor an ecosystems animal population. However, manual inspection of the images produced is expensive, laborious, and time‐consuming. The success of deep learning systems using camera trap images has been previously explored in preliminary stages. These studies, however, are lacking in their practicality. They are primarily focused on extremely large datasets, often millions of images, and there is little to no focus on performance when tasked with species identification in new locations not seen during training. Our goal was to test the capabilities of deep learning systems trained on camera trap images using modestly sized training data, compare performance when considering unseen background locations, and quantify the gradient of lower bound performance to provide a guideline of data requirements in correspondence to performance expectations. We use a dataset provided by Parks Canada containing 47,279 images collected from 36 unique geographic locations across multiple environments. Images represent 55 animal species and human activity with high‐class imbalance. We trained, tested, and compared the capabilities of six deep learning computer vision networks using transfer learning and image augmentation: DenseNet201, Inception‐ResNet‐V3, InceptionV3, NASNetMobile, MobileNetV2, and Xception. We compare overall performance on “trained” locations where DenseNet201 performed best with 95.6% top‐1 accuracy showing promise for deep learning methods for smaller scale research efforts. Using trained locations, classifications with <500 images had low and highly variable recall of 0.750 ± 0.329, while classifications with over 1,000 images had a high and stable recall of 0.971 ± 0.0137. Models tasked with classifying species from untrained locations were less accurate, with DenseNet201 performing best with 68.7% top‐1 accuracy. Finally, we provide an open repository where ecologists can insert their image data to train and test custom species detection models for their desired ecological domain.     

A chaotic neural network mimicking an olfactory system and its application on image recognition

1 Introduction A biological neural system is complicated and ef-ficient. People have tried for years to simulate it to per-form complex signal processing functions. For example,the artificial neural network is a kind of model derivedfrom a biological neural system. Most artificial neuralnetworks simulate some important features such as thethreshold behaviour and plasticity of synapses. However,they are primary simulations and still much simpler incomparison with specific biological neural…     

Wildlife surveillance using deep learning methods

相似文献   

Noise suppressed,multifocus image fusion for enhanced intraoperative navigation

Current intraoperative imaging systems are typically not able to provide ‘sharp’ images over entire large areas or entire organs. Distinct structures such as tissue margins or groups of malignant cells are therefore often difficult to detect, especially under low signal‐to‐noise‐ratio conditions. In this report, we introduce a noise suppressed multifocus image fusion algorithm, that provides detailed reconstructions even when images are acquired under sub‐optimal conditions, such is the case for real time fluorescence intraoperative surgery. The algorithm makes use of the Anscombe transform combined with a multi‐level stationary wavelet transform with individual threshold‐based shrinkage. While the imaging system is integrated with a respiratory monitor triggering system, it can be easily adapted to any commercial imaging system. The developed algorithm is made available as a plugin for Osirix. (© 2013 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Preserving image texture while reducing radiation dose with a deep learning image reconstruction algorithm in chest CT: A phantom study

PurposeTo assess whether a deep learning image reconstruction algorithm (TrueFidelity) can preserve the image texture of conventional filtered back projection (FBP) at reduced dose levels attained by ASIR-V in chest CT.MethodsPhantom images were acquired using a clinical chest protocol (7.6 mGy) and two levels of dose reduction (60% and 80%). Images were reconstructed with FBP, ASIR-V (50% and 100% blending) and TrueFidelity (low (DL-L), medium (DL-M) and high (DL-H) strength). Noise (SD), noise power spectrum (NPS) and task-based transfer function (TTF) were calculated. Noise texture was quantitatively compared by computing root-mean-square deviations (RMSD) of NPS with respect to FBP. Four experienced readers performed a contrast-detail evaluation. The dose reducing potential of TrueFidelity compared to ASIR-V was assessed by fitting SD and contrast-detail as a function of dose.ResultsDL-M and DL-H reduced noise and NPS area compared to FBP and 50% ASIR-V, at all dose levels. At 7.6 mGy, NPS of ASIR-V 50/100% was shifted towards lower frequencies (f_peak = 0.22/0.13 mm⁻¹, RMSD = 0.14/0.38), with respect to FBP (f_peak = 0.30 mm⁻¹). Marginal difference was observed for TrueFidelity: f_peak = 0.33/0.30/0.30 mm⁻¹ and RMSD = 0.03/0.04/0.07 for L/M/H strength. Values of TTF_50% were independent of DL strength and higher compared to FBP and ASIR-V, at all dose and contrast levels. Contrast-detail was highest for DL-H at all doses. Compared to 50% ASIR-V, DL-H had an estimated dose reducing potential of 50% on average, without impairing noise, texture and detectability.ConclusionsTrueFidelity preserves the image texture of FBP, while outperforming ASIR-V in terms of noise, spatial resolution and detectability at lower doses.     

An ensemble learning integration of multiple CNN with improved vision transformer models for pest classification

Pests are the main threats to crop growth, and the precision classification of pests is conducive to formulating effective prevention and governance strategies. In response to the problems of low efficiency and inadaptability to the large-scale environment of existing pest classification methods, this paper proposes a new pest classification method based on a convolutional neural network (CNN) and an improved Vision Transformer model. First, the MMAlNet is designed to extract the characteristics of the identification object from different scales and finer granularity. Then, a classification model called DenseNet Vision Transformer (DNVT) combining a CNN and an improved vision transformer model is proposed. The proposed DNVT captures both long distance dependencies and local characteristic modelling capabilities, which can effectively improve pest classification accuracy. Finally, the ensemble learning algorithm is used to learn MMAlNet and DNVT classification forecasts for soft voting, further enhancing the classification accuracy of pests. The simulation experiment results on the D0 and IP102 datasets show that the proposed method attained a maximum classification of 99.89 and 74.20%, respectively, which is better than other state-of-the-art methods and has a high practical application value.