A systematic evaluation of computation methods for cell segmentation.

Cell segmentation is a fundamental task in analyzing biomedical images. Many computational methods have been developed for cell segmentation, but their performances are not well understood in various scenarios. We systematically evaluated the performance of 18 segmentation methods to perform cell nuclei and whole cell segmentation using light microscopy and fluorescence staining images. We found that general-purpose methods incorporating the attention mechanism exhibit the best overall performance. We identified various factors influencing segmentation performances, including training data and cell morphology, and evaluated the generalizability of methods across image modalities. We also provide guidelines for choosing the optimal segmentation methods in various real application scenarios. We developed Seggal, an online resource for downloading segmentation models already pre-trained with various tissue and cell types, which substantially reduces the time and effort for training cell segmentation models.

From Ensemble Clustering to Subspace Clustering: Cluster Structure Encoding.

In this study, we propose a novel algorithm to encode the cluster structure by incorporating ensemble clustering (EC) into subspace clustering (SC). First, the low-rank representation (LRR) is learned from a higher order data relationship induced by ensemble K-means coding, which exploits the cluster structure in a co-association matrix of basic partitions (i.e., clustering results). Second, to provide a fast predictive coding mechanism, an encoding function parameterized by neural networks is introduced to predict the LRR derived from partitions. These two steps are jointly proceeded to seamlessly integrate partition information and original features and thus deliver better representations than the ones obtained from each single source. Moreover, an alternating optimization framework is developed to learn the LRR, train the encoding function, and fine-tune the higher order relationship. Extensive experiments on eight benchmark datasets validate the effectiveness of the proposed algorithm on several clustering tasks compared with state-of-the-art EC and SC methods.

Adversarial Multiview Clustering Networks With Adaptive Fusion.

The existing deep multiview clustering (MVC) methods are mainly based on autoencoder networks, which seek common latent variables to reconstruct the original input of each view individually. However, due to the view-specific reconstruction loss, it is challenging to extract consistent latent representations over multiple views for clustering. To address this challenge, we propose adversarial MVC (AMvC) networks in this article. The proposed AMvC generates each view's samples conditioning on the fused latent representations among different views to encourage a more consistent clustering structure. Specifically, multiview encoders are used to extract latent descriptions from all the views, and the corresponding generators are used to generate the reconstructed samples. The discriminative networks and the mean squared loss are jointly utilized for training the multiview encoders and generators to balance the distinctness and consistency of each view's latent representation. Moreover, an adaptive fusion layer is developed to obtain a shared latent representation, on which a clustering loss and the l1,2 -norm constraint are further imposed to improve clustering performance and distinguish the latent space. Experimental results on video, image, and text datasets demonstrate that the effectiveness of our AMvC is over several state-of-the-art deep MVC methods.

Role of metabolic equivalent between calcium intake and vertebral fractures: a cross-sectional study of NHANES 2013-2014.

BACKGROUND: This study was to analyze the association of calcium intake and metabolic equivalent (MET) with vertebral fractures, and to explore the role of MET between calcium intake and vertebral fractures. METHOD: This cross-sectional study used data from the National Health and Nutrition Examination Surveys (NHANES) 2013-2014. The study involved individuals aged ≥ 50 years old with complete information on vertebral fracture, calcium intake, and physical activity. Vertebral fracture assessment is obtained using dual-energy x-ray absorptiometry to perform a lateral scan of the thoracolumbar spine. Calcium intake included total nutrient intake and total dietary supplements. The total MET is the sum of the METs for each activity (Vigorous/ moderate work-related activities, walking or bicycling for transportation and vigorous/ moderate recreational activities). Univariate and multivariate logistic regression analyses were utilized to investigate the effect of calcium intake, MET, and their combined effect on vertebral fracture. RESULTS: A total of 766 participants were included in the analysis, and 54 participants had vertebral fractures. The median calcium intake and MET were 8.43 mcg and 280.00, respectively. Multivariate results showed that neither calcium intake nor MET as continuous or categorical variables was significantly associated with vertebral fractures. MET < 160 and calcium intake ≥ 670 mg group was associated with the decreased risks of vertebral fracture [odds ratio (OR) = 0.47, 95% confidence interval (CI): 0.26-0.83, P = 0.032] after adjusting for age, race, energy, total femur bone mineral density (BMD), and femoral neck BMD. In the group of MET < 160, increased calcium intake was associated with a reduced risk of vertebral fracture, with a decreased OR value. In the group of MET ≥ 160, increased calcium intake was associated with an increased risk of vertebral fracture, with an increased OR value. CONCLUSION: The combination of MET < 160 and calcium intake ≥ 670 mg was associated with decreased risks of vertebral fractures. There may be an interaction between calcium intake and MET on vertebral fracture risk.

Multi-View Subspace Clustering via Structured Multi-Pathway Network.

Recently, deep multi-view clustering (MVC) has attracted increasing attention in multi-view learning owing to its promising performance. However, most existing deep multi-view methods use single-pathway neural networks to extract features of each view, which cannot explore comprehensive complementary information and multilevel features. To tackle this problem, we propose a deep structured multi-pathway network (SMpNet) for multi-view subspace clustering task in this brief. The proposed SMpNet leverages structured multi-pathway convolutional neural networks to explicitly learn the subspace representations of each view in a layer-wise way. By this means, both low-level and high-level structured features are integrated through a common connection matrix to explore the comprehensive complementary structure among multiple views. Moreover, we impose a low-rank constraint on the connection matrix to decrease the impact of noise and further highlight the consensus information of all the views. Experimental results on five public datasets show the effectiveness of the proposed SMpNet compared with several state-of-the-art deep MVC methods.

Large-Scale Subspace Clustering by Independent Distributed and Parallel Coding.

Subspace clustering is a popular method to discover underlying low-dimensional structures of high-dimensional multimedia data (e.g., images, videos, and texts). In this article, we consider a large-scale subspace clustering (LS2C) problem, that is, partitioning million data points with a millon dimensions. To address this, we explore an independent distributed and parallel framework by dividing big data/variable matrices and regularization by both columns and rows. Specifically, LS2C is independently decomposed into many subproblems by distributing those matrices into different machines by columns since the regularization of the code matrix is equal to a sum of that of its submatrices (e.g., square-of-Frobenius/ l1 -norm). Consensus optimization is designed to solve these subproblems in a parallel way for saving communication costs. Moreover, we provide theoretical guarantees that LS2C can recover consensus subspace representations of high-dimensional data points under broad conditions. Compared with the state-of-the-art LS2C methods, our approach achieves better clustering results in public datasets, including a million images and videos.

Learnable Subspace Clustering.

This article studies the large-scale subspace clustering (LS2C) problem with millions of data points. Many popular subspace clustering methods cannot directly handle the LS2C problem although they have been considered to be state-of-the-art methods for small-scale data points. A simple reason is that these methods often choose all data points as a large dictionary to build huge coding models, which results in high time and space complexity. In this article, we develop a learnable subspace clustering paradigm to efficiently solve the LS2C problem. The key concept is to learn a parametric function to partition the high-dimensional subspaces into their underlying low-dimensional subspaces instead of the computationally demanding classical coding models. Moreover, we propose a unified, robust, predictive coding machine (RPCM) to learn the parametric function, which can be solved by an alternating minimization algorithm. Besides, we provide a bounded contraction analysis of the parametric function. To the best of our knowledge, this article is the first work to efficiently cluster millions of data points among the subspace clustering methods. Experiments on million-scale data sets verify that our paradigm outperforms the related state-of-the-art methods in both efficiency and effectiveness.

Impacts of Nitrogen Deficiency on Wheat (Triticum aestivum L.) Grain During the Medium Filling Stage: Transcriptomic and Metabolomic Comparisons.

Nitrogen (N) supplementation is essential to the yield and quality of bread wheat (Triticum aestivum L.). The impact of N-deficiency on wheat at the seedling stage has been previously reported, but the impact of distinct N regimes applied at the seedling stage with continuous application on filling and maturing wheat grains is lesser known, despite the filling stage being critical for final grain yield and flour quality. Here, we compared phenotype characteristics such as grain yield, grain protein and sugar quality, plant growth, leaf photosynthesis of wheat under N-deficient and N-sufficient conditions imposed prior to sowing (120 kg/hm2) and in the jointing stage (120 kg/hm2), and then evaluated the effects of this continued stress through RNA-seq and GC-MS metabolomics profiling of grain at the mid-filling stage. The results showed that except for an increase in grain size and weight, and in the content of total sugar, starch, and fiber in bran fraction and white flour, the other metrics were all decreased under N-deficiency conditions. A total of 761 differentially expressed genes (DEGs) and 77 differentially accumulated metabolites (DAMs) were identified. Under N-deficiency, 51 down-regulated DEGs were involved in the process of impeding chlorophyll synthesis, chloroplast development, light harvesting, and electron transfer functions of photosystem, which resulted in the SPAD and Pn value decreased by 32 and 15.2% compared with N-sufficiency, inhibited photosynthesis. Twenty-four DEGs implicated the inhibition of amino acids synthesis and protein transport, in agreement with a 17-42% reduction in ornithine, cysteine, aspartate, and tyrosine from metabolome, and an 18.6% reduction in grain protein content. However, 14 DEGs were implicated in promoting sugar accumulation in the cell wall and another six DEGs also enhanced cell wall synthesis, which significantly increased fiber content in the endosperm and likely contributed to increasing the thousands-grain weight (TGW). Moreover, RNA-seq profiling suggested that wheat grain can improve the capacity of DNA repair, iron uptake, disease and abiotic stress resistance, and oxidative stress scavenging through increasing the content levels of anthocyanin, flavonoid, GABA, galactose, and glucose under N-deficiency condition. This study identified candidate genes and metabolites related to low N adaption and tolerance that may provide new insights into a comprehensive understanding of the genotype-specific differences in performance under N-deficiency conditions.

Geometrically Editable Face Image Translation With Adversarial Networks.

Recently, image-to-image translation has received increasing attention, which aims to map images in one domain to another specific one. Existing methods mainly solve this task via a deep generative model that they focus on exploring the bi-directional or multi-directional relationship between specific domains. Those domains are often categorized by attribute-level or class-level labels, which do not incorporate any geometric information in learning process. As a result, existing methods are incapable of editing geometric contents during translation. They also neglect to utilize higher-level and instance-specific information to further guide the training process, leading to a great deal of unrealistic synthesized images of low fidelity, especially for face images. To address these challenges, we formulate the general image translation problem as multi-domain mappings in both geometric and attribute directions within an image set that shares a same latent vector. Particularly, we propose a novel Geometrically Editable Generative Adversarial Networks (GEGAN) model to solve this problem for face images by leveraging facial semantic segmentation to explicitly guide its geometric editing. In details, input face images are encoded to their latent representations via a variational autoencoder, a segmentor network is designed to impose semantic information on the generated images, and multi-scale regional discriminators are employed to force the generator to pay attention to the details of key components. We provide both quantitative and qualitative evaluations on CelebA dataset to demonstrate our ability of the geometric modification and our improvement in image fidelity.

Generative Partial Multi-View Clustering With Adaptive Fusion and Cycle Consistency.

Nowadays, with the rapid development of data collection sources and feature extraction methods, multi-view data are getting easy to obtain and have received increasing research attention in recent years, among which, multi-view clustering (MVC) forms a mainstream research direction and is widely used in data analysis. However, existing MVC methods mainly assume that each sample appears in all the views, without considering the incomplete view case due to data corruption, sensor failure, equipment malfunction, etc. In this study, we design and build a generative partial multi-view clustering model with adaptive fusion and cycle consistency, named as GP-MVC, to solve the incomplete multi-view problem by explicitly generating the data of missing views. The main idea of GP-MVC lies in two-fold. First, multi-view encoder networks are trained to learn common low-dimensional representations, followed by a clustering layer to capture the shared cluster structure across multiple views. Second, view-specific generative adversarial networks with multi-view cycle consistency are developed to generate the missing data of one view conditioning on the shared representation given by other views. These two steps could be promoted mutually, where the learned common representation facilitates data imputation and the generated data could further explores the view consistency. Moreover, an weighted adaptive fusion scheme is implemented to exploit the complementary information among different views. Experimental results on four benchmark datasets are provided to show the effectiveness of the proposed GP-MVC over the state-of-the-art methods.

Isolation of novel Lactobacillus with lipolytic activity from the vinasse and their preliminary potential using as probiotics.

Lactobacillus casei f1, L. paracasei f2 and L. paracasei f3 with lipolytic activity were isolated and identified from vinasses according to the morphological-physiological properties detection and 16S rDNA analysis. These three strains showed obvious lipase activities to olive oil and L. casei f1 performed highest enzyme activity of 17.8 U/mL. L. casei f1, L. paracasei f2 and L. paracasei f3 could lipolyze the blending oils, peanut oil and sesame oil with diverse degrading rates. The degrading rates to the preferred oils, L. casei f1 to blending oils, L. paracasei f2 to peanut oil and L. paracasei f3 to sesame oil, were 21.2%, 27.3% and 39.6%, respectively. The corresponding oil degrading rates increased as the cell growth and the highest degrading rates were obtained at the stationary phase with the viable count more than 7.5 LogCFU/mL. By GC-MS analysis, L. casei f1, L. paracasei f2 and L. paracasei f3 performed diverse lipolytic capacities to the 12 kinds of fat acids and all of them preferred to hydrolyze the linoleic acid with the degrading rate of 49.11%, 31.83% and 64.44%, respectively. These three strains showed considerable probiotic properties, displaying higher than 106 CFU/mL desirable viable count though the simulated gastrointestinal tract, as well as inhibiting six indicator bacteria. These results suggested that the three isolated strains could be considered as novel probiotic candidates and applied in the food industry.

Marginalized Multiview Ensemble Clustering.

Multiview clustering (MVC), which aims to explore the underlying cluster structure shared by multiview data, has drawn more research efforts in recent years. To exploit the complementary information among multiple views, existing methods mainly learn a common latent subspace or develop a certain loss across different views, while ignoring the higher level information such as basic partitions (BPs) generated by the single-view clustering algorithm. In light of this, we propose a novel marginalized multiview ensemble clustering (M2VEC) method in this paper. Specifically, we solve MVC in an EC way, which generates BPs for each view individually and seeks for a consensus one. By this means, we naturally leverage the complementary information of multiview data upon the same partition space. In order to boost the robustness of our approach, the marginalized denoising process is adopted to mimic the data corruptions and noises, which provides robust partition-level representations for each view by training a single-layer autoencoder. A low-rank and sparse decomposition is seamlessly incorporated into the denoising process to explicitly capture the consistency information and meanwhile compensate the distinctness between heterogeneous features. Spectral consensus graph partitioning is also involved by our model to make M2VEC as a unified optimization framework. Moreover, a multilayer M2VEC is eventually delivered in a stacked fashion to encapsulate nonlinearity into partition-level representations for handling complex data. Experimental results on eight real-world data sets show the efficacy of our approach compared with several state-of-the-art multiview and EC methods. We also showcase our method performs well with partial multiview data.

Adversarial Action Prediction Networks.

Different from after-the-fact action recognition, action prediction task requires action labels to be predicted from partially observed videos containing incomplete action executions. It is challenging because these partial videos have insufficient discriminative information, and their temporal structure is damaged. We study this problem in this paper, and propose an efficient and powerful deep network for learning representative and discriminative features for action prediction. Our approach exploits abundant sequential context information in full videos to enrich the feature representations of partial videos. This information is encoded in latent representations using a variational autoencoder (VAE), which are encouraged to be progress-invariant. Decoding such latent representations using another VAE, we can reconstruct missing information in the features extracted from partial videos. An adversarial learning scheme is adopted to differentiate the reconstructed features from the features directly extracted from full videos in order to well align their distributions. A multi-class classifier is also used to encourage the features to be discriminative. Our network jointly learns features and classifiers, and generates the features particularly optimized for action prediction. Extensive experimental results on UCF101, Sports-1M and BIT datasets demonstrate that our approach remarkably outperforms state-of-the-art methods, and shows significant speedup over these methods. Results also show that actions differ in their prediction characteristics; some actions can be correctly predicted even though only the beginning 10% portion of videos is observed.

Multi-View Saliency-Guided Clustering for Image Cosegmentation.

Image cosegmentation aims at extracting the common objects from multiple images simultaneously. Existing methods mainly solve cosegmentation via the pre-defined graph, which lacks flexibility and robustness to handle various visual patterns. Besides, similar backgrounds also confuse the identifying of the common foreground. To address these issues, we propose a novel Multi-view Saliency-Guided Clustering algorithm (MvSGC) for the image cosegmentation task. In our model, the unsupervised saliency prior is used as partition-level side information to guide the foreground clustering process. To achieve robustness to noises and missing observations, similarities on instance-level and partition-level are both considered. Specifically, a unified clustering model with cosine similarity is proposed to capture the intrinsic structure of data and keep partition result consistent with the side information. Moreover, we leverage multi-view weight learning to integrate multiple feature representations to further improve the robustness of our approach. A K-means-like optimization algorithm is developed to proceed the constrained clustering in a highly efficient way with theoretical support. Experimental results on three benchmark datasets (i.e., the iCoseg, MSRC and Internet image dataset) and one RGB-D image dataset demonstrate the superiority of applying our clustering method for image cosegmentation.

The relationship between serum fibrinogen-like protein 2 concentrations and 30-day mortality of patients with traumatic brain injury.

BACKGROUND: Fibrinogen-like protein 2 (FGL2) is an inflammatory procoagulant protein. We discerned the impact of serum FGL2 on trauma severity and 30-day mortality in patients with traumatic brain injury (TBI). METHODS: A total of 114 severe TBI patients were subjected to assessment of trauma severity using the Glasgow coma scale (GCS). Measurement of the serum concentrations of FGL2 was done. 114 matched control subjects for their age and sex were included for comparison of serum concentration of FGL2. RESULTS: The concentration of FGL2 was dramatically increased in the patients as compared with the control subjects. FGL2 concentration was inversely correlated with GCS score among the patients. The non-survivors within 30 days exhibited substantially higher FGL2 concentrations than the alive. FGL2 concentrations discriminated the patients at risk of 30-day death with significantly high area under receiver operating characteristic curve. Serum FGL2 emerged as an independent predictor for mortality and overall survival at 30 days after head trauma. CONCLUSIONS: Serum FGL2 is a promising biomarker for assessing the severity and prognosis in severe TBI.

The Pathogenesis of Subacute Subdural Hematoma: A Report of 3 Cases and Literature Review.

OBJECTIVE: To discuss the pathologic mechanism of subacute subdural hematoma (sASDH). METHODS: Three typical cases of sASDH were reported, and related literature in Chinese published in the past 15 years was reviewed. RESULTS: Intervals from onset of acute subdural hematoma to surgery or symptom deterioration resulting in sASDH were 12.5-15.5 days (mean 14.1 days). Delayed liquefaction of hematoma clots occurred in all 3 reported cases. One patient achieved good curative effect after administration of dexamethasone, and another patient relapsed owing to poor drainage after evacuation of hematoma. CONCLUSIONS: The conversion of acute subdural hematoma to sASDH is an inflammatory reaction process with very regular in time, and it is speculated that the pathologic mechanism may be a delayed hypersensitivity reaction. Antigen released during the liquefaction process of blood clot, with subdural neomembrane cells as antigen-presenting cells, is presented to the T lymphocytes released from the capillaries in the neomembrane and forms sensitized T lymphocytes. When the subsequent antigen is released from the blood clots with a delayed liquefaction and is exposed to sensitized T lymphocytes, the delayed hypersensitivity process occurs.

Subdural Abnormal Communicating Vessel in Chronic Subdural Hematoma.

Traditionally, lacerations of bridging vessels were surmised to cause chronic subdural hematoma (CSDH), although neither observation studies nor medical research was able to testify this. Nowadays, an inflammatory process is known to take place in the development of CSDH. Of note, post-traumatic angiogenesis at its early stage also features inflammation with immune cell infiltration. The authors found a patient suffering from CSDH with unusual angiogenesis between dura and pia matters. The observation of dura-and-pia angiogenesis may be a piece of evidence to underline compensatory reaction of central nervous system to offset the negative effects produced by CSDH, and points out to a possible approach of bolstering angiogenesis to manage ischemic diseases in cerebral hemispheres.

Partition Level Constrained Clustering.

Constrained clustering uses pre-given knowledge to improve the clustering performance. Here we use a new constraint called partition level side information and propose the Partition Level Constrained Clustering (PLCC) framework, where only a small proportion of the data is given labels to guide the procedure of clustering. Our goal is to find a partition which captures the intrinsic structure from the data itself, and also agrees with the partition level side information. Then we derive the algorithm of partition level side information based on K-means and give its corresponding solution. Further, we extend it to handle multiple side information and design the algorithm of partition level side information for spectral clustering. Extensive experiments demonstrate the effectiveness and efficiency of our method compared to pairwise constrained clustering and ensemble clustering methods, even in the inconsistent cluster number setting, which verifies the superiority of partition level side information to pairwise constraints. Besides, our method has high robustness to noisy side information, and we also validate the performance of our method with multiple side information. Finally, the image cosegmentation application based on saliency-guided side information demonstrates the effectiveness of PLCC as a flexible framework in different domains, even with the unsupervised side information.

[Spectral Characteristics of Spring Maize Varieties with Different Heat Tolerance to High Temperature].

This paper discussed the response of spectral characteristics on high temperature at grain filling stage of different spring maize varieties by adopting two spectrometer (SPAD-502 Chlorophyll Meter and Sunscan Plant Canopy Analyzer), and analyzed the impact of high temperature on the photosynthetic properties of spring maize in North China Plain. The test was conductedfrom the year 2011 to 2012 in Wuqiao County, Hebei Province. This test chose three different varieties, i. e. Tianyu 198 (TY198), Xingyu 998 (XY998) and Tianrun 606 (TR606), then two sowing date (April 15th and April 25th) was set. We analyzed chlorophyll relative content (SPAD), leaf area index (LAI) and photosynthetically active radiation (PAR) at grain filling stage. The results showed that the days of daily maximum temperature above 33 °C and the mean day temperature at grain filling stage in spring maize sowing on April 15th increased 3.5 d and 0.8 °C, respectively, compared to that sowing on April 25th, moreover the sunshine hours, rainfall, diurnal temperature and length of growing period were similar. Compared with XY998 and TR606, TY198's stress tolerance indices (STI) increased by 2.9% and 11.0%, respectively. According to STI from high to low order, TY198, XY998 and TR606 respectively as heat resistant type, moderate heat resistant type and thermo-labile type variety. TY198, compared with XY998 and TR606 sowing on April 15th, yield increased by 4.1% and 13.7%, SPAD increased by 12.5% and 19.6%, LAI increased by 5.3% and 5.6%, PAR increased by 4.0% and 14.0%. Sowing on April 15th, yield increased by 1.3% and 2.8%, SPAD increased by 3.5% and 6.0%, LAI increased by 1.7% and 4.1%, PAR increased by -4.4% and 0.9%. Three varieties had significant yield differences in the environment of high temperature stress, heat resistant type have significant (p < 0.05) advantage in the aspect of yield, SPAD and LAI. The production of TY198, XY998 and TR606 sowing on April 15th compared to that sowing on April 25th decreased by 3.2%, 5.9% and 12.6%, and SPAD decreased by 8.6%, 12.4% and 15.7%, LAI decreased by 11.7%, 17.6% and 19.8%, PAR decreased by 3.4%, 11.3% and 14.5%; STI had a significant negatively correlated with SPAD fall range (r = -0.883, p < 0.05) and LAI fall range (r = -0.853, P < 0.05), and highly significantly negatively correlated with PAR fall range (r = -0.923, p < 0.01); while SPAD fall range and PAR fall range showed a significant positive correlation (r = 0.872, p < 0.05); LAI fall range and PAR fall range were significantly positive correlation (r = 0.943, p < 0.05). In conclusion, heat tolerant type varieties of spring maize under high temperature stress at gain filling stage could maintain a relatively high content of chlorophyll at the individual level, a relatively high leaf area at the group level, and then keep a higher luminous energy interception and utilization, and weakened inhibition magnitude of high temperature on photosynthetic capacity, reduced the yield fall range, then achieved high and stable yield. The heat tolerance in varieties could be one of the main indicators for identification and evaluation the response to high temperature by spectral characteristics (SPAD, LAI and PAR). Thus it provides a basis by using spectral characteristics to study heat tolerance on maize.

[Establishment of The Crop Growth and Nitrogen Nutrition State Model Using Spectral Parameters Canopy Cover].

In order to explore a non-destructive monitoring technique, the use of digital photo pixels canopy cover (CC) diagnosis and prediction on maize growth and its nitrogen nutrition status. This study through maize canopy digital photo images on relationship between color index in the photo and the leaf area index (LAI), shoot dry matter weight (DM), leaf nitrogen content percentage (N%). The test conducted in the Chinese Academy of Agricultural Science from 2012 to 2013, based on Maize canopy Visual Image Analysis System developed by Visual Basic Version 6.0, analyzed the correlation of CC, color indices, LAI, DM, N% on maize varieties (Zhongdan909, ZD 909) under three nitrogen levels treatments, furthermore the indicators significantly correlated were fitted with modeling, The results showed that CC had a highly significant correlation with LAI (r = 0.93, p < 0.01), DM (r = 0. 94, p < 0.01), N% (r = 0.82, p < 0.01). Estimating the model of LAI, DM and N% by CC were all power function, and the equation respectively were y = 3.281 2x(0.763 9), y = 283.658 1x(0.553 6) and y = 3.064 5x(0.932 9); using independent data from modeling for model validation indicated that R2, RMSE and RE based on 1 : 1 line relationship between measured values and simulated values in the model of CC estimating LAI were 0.996, 0.035 and 1.46%; R2, RMSE and RE in the model of CC estimating DM were 0.978, 5.408 g and 2.43%; R2, RMSE and RE in the model of CC estimating N% were 0.990, 0.054 and 2.62%. In summary, the model can comparatively accurately estimate the LAI, DM and N% by CC under different nitrogen levels at maize grain filling stage, indicating that it is feasible to apply digital camera on real-time undamaged rapid monitoring and prediction for maize growth conditions and its nitrogen nutrition status. This research finding is to be verified in the field experiment, and further analyze the applicability throughout the growing period in other maize varieties and different planting density.