Research article Topical Sections

Microbiological safety of flours used in follow up for infant formulas produced in Ouagadougou, Burkina Faso

  • The prevalence of diarrheal diseases in children aged from 6 to 24 months in Burkina Faso is 38%. These diarrheas may be due to the consumption of contaminated weaning food. Therefore, the microbiological quality of follow up infant flours used as supplement foods is a key-point to reduce children diseases. In this study, the microbiological safety of locally-produced infant flours was investigated. One hundred and ninety-nine (199) samples were collected mainly in retails outlets and in Recovery and Nutrition Education Centers. According to the Burkina Faso regulations, microbiological analyses were carried out for Total Aerobic Mesophilic Flora (TAMF), thermotolerant coliforms, Salmonella spp. and yeasts/molds. The bacterial and fungal isolates were identified using phenotypic and genotypic methods and the study of the production of mycotoxins was carried out from the fungal isolates. In collected samples, the TAMF count ranged from 0 to 1.8 × 106 CFU/g with a total average of 6.3 × 104 CFU/g. About 2% of the samples had a microbial load exceeding the standards (105 CFU/g). No Salmonella spp. was isolated in the final infant flours. However, the presence of Enterobacteriaceae (Klebsiella spp. Enterobacter spp. and Cronobacter spp.) was detected and molecular characterization revealed also the presence of fungal species of the genus Aspergillus spp., Penicillium spp. and Fusarium spp. Some of these species were found to produce aflatoxins, ochratoxin A and fumonisins, which are potential carcinogenic toxins. These results demonstrated the need for a preventive approach based on the application of Hazard Analysis Critical Control Point in the food industry to ensure food safety of infant flours in Burkina Faso.

    Citation: Larissa Y. Waré, Augustini P. Nikièma, Jean C. Meile, Saïdou Kaboré, Angélique Fontana, Noël Durand, Didier Montet, Nicolas Barro. Microbiological safety of flours used in follow up for infant formulas produced in Ouagadougou, Burkina Faso[J]. AIMS Microbiology, 2018, 4(2): 347-361. doi: 10.3934/microbiol.2018.2.347

    Related Papers:

    [1] Chiara Ceresa, Francesco Tessarolo, Devid Maniglio, Iole Caola, Giandomenico Nollo, Maurizio Rinaldi, Letizia Fracchia . Inhibition of Candida albicans biofilm by lipopeptide AC7 coated medical-grade silicone in combination with farnesol. AIMS Bioengineering, 2018, 5(3): 192-208. doi: 10.3934/bioeng.2018.3.192
    [2] Chiara Ceresa, Maurizio Rinaldi, Letizia Fracchia . Synergistic activity of antifungal drugs and lipopeptide AC7 against Candida albicans biofilm on silicone. AIMS Bioengineering, 2017, 4(2): 318-334. doi: 10.3934/bioeng.2017.2.318
    [3] Amreeta Sarjit, Sin Mei Tan, Gary A. Dykes . Surface modification of materials to encourage beneficial biofilm formation. AIMS Bioengineering, 2015, 2(4): 404-422. doi: 10.3934/bioeng.2015.4.404
    [4] Ana Meireles, Ana L. Gonçalves, Inês B. Gomes, Lúcia Chaves Simões, Manuel Simões . Methods to study microbial adhesion on abiotic surfaces. AIMS Bioengineering, 2015, 2(4): 297-309. doi: 10.3934/bioeng.2015.4.297
    [5] Merily Horwat, Meggie Tice, Birthe V. Kjellerup . Biofilms at work: Bio-, phyto- and rhizoremediation approaches for soils contaminated with polychlorinated biphenyls. AIMS Bioengineering, 2015, 2(4): 324-334. doi: 10.3934/bioeng.2015.4.324
    [6] Sara R Robertson, Robert JC McLean . Beneficial biofilms. AIMS Bioengineering, 2015, 2(4): 437-448. doi: 10.3934/bioeng.2015.4.437
    [7] Yung-Hua Li, Xingxing Huang, Xiao-Lin Tian . Recent advances in dental biofilm: impacts of microbial interactions on the biofilm ecology and pathogenesis. AIMS Bioengineering, 2017, 4(3): 335-350. doi: 10.3934/bioeng.2017.3.335
    [8] Lucie Semenec, Ashley E Franks . Delving through electrogenic biofilms: from anodes to cathodes to microbes. AIMS Bioengineering, 2015, 2(3): 222-248. doi: 10.3934/bioeng.2015.3.222
    [9] Frank B Dazzo, Youssef G Yanni, Ashley Jones, Abdelgawad Y Elsadany . CMEIAS bioimage informatics that define the landscape ecology of immature microbial biofilms developed on plant rhizoplane surfaces. AIMS Bioengineering, 2015, 2(4): 469-486. doi: 10.3934/bioeng.2015.4.469
    [10] Joe A. Lemire, Marc A. Demeter, Iain George, Howard Ceri, Raymond J. Turner . A novel approach for harnessing biofilm communities in moving bed biofilm reactors for industrial wastewater treatment. AIMS Bioengineering, 2015, 2(4): 387-403. doi: 10.3934/bioeng.2015.4.387
  • The prevalence of diarrheal diseases in children aged from 6 to 24 months in Burkina Faso is 38%. These diarrheas may be due to the consumption of contaminated weaning food. Therefore, the microbiological quality of follow up infant flours used as supplement foods is a key-point to reduce children diseases. In this study, the microbiological safety of locally-produced infant flours was investigated. One hundred and ninety-nine (199) samples were collected mainly in retails outlets and in Recovery and Nutrition Education Centers. According to the Burkina Faso regulations, microbiological analyses were carried out for Total Aerobic Mesophilic Flora (TAMF), thermotolerant coliforms, Salmonella spp. and yeasts/molds. The bacterial and fungal isolates were identified using phenotypic and genotypic methods and the study of the production of mycotoxins was carried out from the fungal isolates. In collected samples, the TAMF count ranged from 0 to 1.8 × 106 CFU/g with a total average of 6.3 × 104 CFU/g. About 2% of the samples had a microbial load exceeding the standards (105 CFU/g). No Salmonella spp. was isolated in the final infant flours. However, the presence of Enterobacteriaceae (Klebsiella spp. Enterobacter spp. and Cronobacter spp.) was detected and molecular characterization revealed also the presence of fungal species of the genus Aspergillus spp., Penicillium spp. and Fusarium spp. Some of these species were found to produce aflatoxins, ochratoxin A and fumonisins, which are potential carcinogenic toxins. These results demonstrated the need for a preventive approach based on the application of Hazard Analysis Critical Control Point in the food industry to ensure food safety of infant flours in Burkina Faso.


    Electronic health records (EHRs) contain a wealth of information on individual patient diagnoses, tests, treatments and outcomes, which can be effectively utilized for clinical prediction studies, such as disease-assisted diagnosis and risk prediction. Predicting the risk of mortality in the ICU is a critical step in the treatment of critically ill patients, and if a patient is at high risk of mortality and stays in the ICU for a long time, a lot of medical resources are used and the burden on the patient's family increases. By predicting the risk of mortality for a specific patient, it can help clinicians identify patients whose condition is likely to deteriorate, so that they can take appropriate treatment measures to prevent this group of patients from worsening and reduce the length of ICU stay.

    In recent years, medical research has gradually shifted from a population-based perspective to a personalized perspective, a trend known as precision medicine [1]. Taking inspiration from clinicians, who tend to rely on their past experience in treating similar patients when making clinical judgments, two patients should have similar clinical outcomes if they have similar clinical variables or histological characteristics. Node similarity, based on local information and proposed by some researchers [2], argues that the greater the number of common neighbors that exist between two nodes in a network, the more similar features these two nodes have, i.e., the greater the likelihood that these two nodes belong to the same type of label [3]. However, few people have applied the node similarity problem to patient outcome prediction problems, and according to recent surveys on the application of patient similarity networks in precision medicine or health data processing [4,5,6,7], representing data as graphs has the advantage of being highly interpretable and protecting privacy, because patient-specific information cannot be recovered from the similarity metric. Based on these findings, this paper proposes a patient similarity-based dynamic graphical attention network model for predicting patients' mortality risk. Specifically, the similarity relationship between patients is determined based on their clinical data, such as diagnosis, examination and demographics during hospitalization. The two-by-two similarity between patients can be naturally represented as Graph-Patient Similarity Networks (PSNs). Where each node represents a patient and the similarity between patients calculated using clinical data is represented as a weighted edge. Similar patients often have similar treatment trajectories and outcomes, so information from similar patients can often provide powerful support for outcome prediction, disease risk prediction, etc. [6]. This transforms the patient clinical outcome prediction task into a node classification task.

    The existence of many cases of unmeasured or unrecorded values of patients in electronic medical record data leads to a high number of missing values in the dataset, and most of the current processing of missing values focuses on methods, such as deletion [8], manual filling, filling of global constants such as N/A and Null and filling of mean or average values [9], which require time and effort on the one hand and low precision on the other. The GAT model [10] takes into account the different importance of different neighboring nodes, assigns different weights to different nodes in the neighborhood, and pays attention to the local structural features of the nodes, but also ignores the non-local information of the nodes, such as the higher-order nodes that are most relevant to that node. Additionally, GAT can only calculate static attention, which has the same attention level for each node, and severely impairs and limits the expressiveness of the attention mechanism [11].

    In order to solve the above problems, we propose a Patient Similarity Dynamic Graph Attention Networks (PS-DGAT) based model for mortality risk prediction in ICU, which first constructs a patient similarity network by calculating the similarity of patient clinical data to represent similarities between patients, then fills in the missing features based on the data of neighboring patients in the network and update the patient similarity network, and finally predicts the risk of mortality of patients by the dynamic GAT model.

    The major contributions made in this paper are as follows:

    1) We construct a patient similarity network by calculating the similarity of patient clinical data and transforming the structured data into a graph structure, thus representing the relationship between patients more clearly.

    2) We use the data of adjacent patients in the network to complete the missing data and reconstruct the patient similarity network.

    3) We propose a PS-DGAT model to predict the mortality of ICU patients, and prioritize the vectors with large weights by adjusting the operation order of weighted vectors in GAT to make them better than GAT.

    4) We conduct a large number of experiments on the international large-scale public data set MIMIC-III, and compare with the current popular methods to confirm the effectiveness of the proposed model.

    Mortality risk prediction is an important task in the medical field and has attracted the attention of an increasing number of researchers in recent years. For example, some researchers [12,13] used traditional machine learning algorithms, such as support vector machines and logistic regression to predict the risk of mortality in heart failure patients, but these methods ignore information about the similarity between patients, resulting in limited prediction performance. To overcome these problems, some researchers have used graph neural networks to predict the risk of mortality in heart failure patients. For example, Lu et al. [14] proposed a graph neural network-based creation of weighted patient networks to extract potential relationships between patients for chronic disease prediction by projecting patient-disease dichotomous graphs. Since traditional GNNs cannot adequately consider the variability of neighboring nodes, the structural information learned is not comprehensive and robust enough [15]. Some other researchers [16] used the GAT model to predict the mortality risk of heart failure patients, achieving adaptive aggregation of different nodes. However, GAT can only calculate static attention, and for each query node, the attention level is the same, which limits its expressive power.

    In recent years, an increasing number of researchers have begun to use patient similarity for healthcare data analysis and prediction. Patient similarity analysis simulates the thought process of senior physicians comparing patients, and it refers to selecting clinical concepts (such as diagnosis, symptoms, examination tests, family history, past history, exposure environment, drugs, surgery, genes, etc.) as the characteristic terms of patients in a specific medical setting, quantitatively analyzing the distance between concepts in the semantic space of complex concepts, thus dynamically measuring the distance between patients and screening out the index patient similarity groups [17], so as to assess the current status of patients, predict their prognosis and recommend treatment options. Common similarity calculation methods include Euclidean distance, cosine similarity, Pearson correlation coefficient, etc. Some researchers have constructed similarity networks by calculating the similarity between patients, for example, in disease prediction, drug treatment, etc. [18,19]. However, these methods calculate patient similarity based on only a single feature or a few features, without considering the combination of patients on multiple features, so it is difficult to reflect the similarity relationship between patients comprehensively and accurately. Moreover, these methods often ignore the missing data, which may lead to inaccurate similarity calculation and thus affect the accuracy of prediction results.

    In summary, traditional machine learning methods usually ignore the problem of similarity between patients when predicting their mortality risk. Also, the processing methods for missing values are simpler and cannot fully utilize the clinical data information of similar patients. In addition, despite the rapid development of deep learning in recent years, there is a lack of research in clinical risk prediction using graph neural networks. In this paper, we propose PS-DGAT for predicting the risk of mortality of patients in the ICU. The model uses several features, such as demographic characteristics, laboratory tests and comorbidities to calculate the similarity between patients, and represent the similarity relationship between patients as a patient similarity network. Missing features are filled in by using data from neighbor patients in the network. Finally, patient mortality risk prediction was achieved using a dynamic graph attention network, thus improving the accuracy of the prediction model.

    The detailed architecture of the PS-DGAT model is shown in Figure 1.

    Figure 1.  Model flow chart.

    The model consists of the following three modules. 1) Data extraction module, which extracts patient feature data from the MIMIC-III database and screens the data according to the screening criteria. 2) Data processing module, which in the first step preprocesses the extracted data by filling in the missing values and then standardizing them, and then calculates the patient similarity and establishes the connecting edges between patient nodes whose similarity is greater than 0.5, thus constructing a patient similarity network. Then based on patient similarity the filled missing data is processed again by the feature complementation method, which reconstructs the missing features are reconstructed by propagating the known features on the graph. 3) The graph neural network module: The module mainly uses the dynamic attention mechanism to extract features from the patient similarity network reconstructed after feature completion, learns the structural features of the nodes, and obtains the vector representation of the patient nodes in the low-dimensional embedding space to realize the prediction of patient mortality risk. The following will introduce these three modules and the training process of the PS-DGAT model in detail.

    Similarity of binary features. Gender and comorbidity features are 0, 1 binary features, which can form a binary feature set respectively. A and B represent two patients respectively, X represents the value of a feature corresponding to the patient and the similarity of patients can be calculated according to Eq (1), where the Sbf(A, B) value consistency is 1, and value inconsistency is 0.

    $ {S}_{\text{bf}}(A, B) = \left\{
    \right. $
    (1)

    Similarity of numerical features. Age, BMI, laboratory tests and vital signs are numerical features, which do not involve the time axis after normal normalization, respectively, and we form a 41-dimensional vector to calculate the similarity of numerical features.

    $ {S}_{\text{nf}}(A, B) = {\sqrt{{\displaystyle {\sum }_{\text{i}}(XA\text{i}+XB\text{i}{)}^{2}}}}^{} $ (2)

    where $ {X_{A{\text{i}}}} $ and $ {X_{B{\text{i}}}} $ denote the value of the i-th feature of the patient, i = 1, 2, ..., 41.

    According to the degree of influence of each feature on the patient's outcome, the similarity measure between the two patients was obtained after weighted summation. The calculation formula is as follows:

    $ S(A, B) = {w}_{1}·{S}_{bf}+{w}_{2}·{S}_{nf} $ (3)

    $ {w}_{i} $ is the weight of feature similarity, satisfying $ {w}_{i}\in \left[\mathrm{0, 1}\right] $, $ {\sum }_{i}^{2}{w}_{i} = 1 $. Obviously, $ S(A, B)\in \left[\mathrm{0, 1}\right] $, and the closer to 1, the higher the degree of similarity between patients. Referring to the current relevant research, this paper sets the weights of binary features and numerical features not 0.4 and 0.6 respectively.

    After using the above method to calculate the similarity of the patient nodes, we obtain a similarity matrix, which represents the similarity score between each pair of patient nodes. Due to the large data latitude, we use the sampling method to select the data points with a large number of edges from the high-dimensional similarity matrix as the seed nodes, and use the random walk to sample the probability from the current node according to the weight, select the next node as the sampling result and update the current node as a new node. Repeat this step until a sufficient number of data points are sampled. In order to clearly show the similarity between patients, we sampled 20 data points and plotted the corresponding heat map, label code per cell per patient number.

    Figure 2.  Similarity matrix heat map.

    Through the sampled heat map, we can observe the similar patterns and cluster structures between different patients. Although we may lose some details and local information during the sampling process, the simplified heat map still provides meaningful information about the overall similarity relationship. This can help to understand the similarity between patients and provide intuitive visualization.

    Through the analysis of hyperparameters, we set the similarity threshold to 0.5, and add an edge between two nodes greater than or equal to 0.5 in the similarity matrix. In this way, we get a weighted patient similarity network. In order to visualize the patient similarity network more intuitively, we use the network diagram to display the 20 important nodes sampled above. As shown in Figure 3, each node represents a patient, and the edge represents the similarity between patients.

    Figure 3.  Patient similarity network graph.

    To address the problem of serious missing data, since in node similarity neighboring nodes tend to have similar feature vectors [3], we propose to complete the node features using a patient similarity-based feature completion method. Specifically, the missing features are updated by iteratively propagating the known features of neighboring nodes on the graph to better utilize the relationships between neighboring nodes to update in the missing feature values, resulting in a complete feature matrix. We then recalculated the patient similarity matrix using the patched feature data, which in turn reconstructed the patient similarity network.

    Feature completion is aimed at the problem of missing node attributes, while the dataset in this paper has the problem of non-existent edge structure in addition to missing attributes. Therefore, the missing data are first filled using KNN, respectively, to add labels to the missing data, and construct the similarity network using the method in Section 3.1. Then, the feature complementation algorithm reconstructs the missing features by iteratively diffusing the known features in the graph. Finally, the patient similarity network is reconstructed with updated node features, which are then fed into the downstream GNN model, and then generates predictions. The propagation process of the algorithm is implemented through the following steps:

    Figure 4.  Feature completion framework.

    Initialization. Assignment of known feature values (i.e., non-missing values) to the corresponding nodes.

    Propagation. The missing values are updated by calculating the similarity between neighboring nodes, specifically, the updated values of the node feature vectors are calculated by the following equation:

    $ xi(t + 1) = \frac{{sum(Sij*xj(t))}}{{sum(Sij)}} $ (4)

    where $ {x}_{i}(t+1) $ denotes the feature vector of node i at the t-th iteration; $ {S}_{ij} $ denotes the similarity between node i and node j; $ {x_j}(t) $ denotes the feature vector of node $ j $ at the t-th iteration; $ sum({S_{ij}}) $ denotes the similarity sum between neighboring nodes.

    End condition. Stop feature propagation iteration when the feature vector's no longer changes or changes slightly or reaches the maximum number of iterations (40 times).

    Output result. Output the node feature matrix $ {X^{'}} $ after convergence of iterations, where the feature vector of node j with missing features is reconstructed.

    In summary, the flow of the patient similarity-based feature completion algorithm is as follows:

    Algorithm 1. Feature Completion
    Input: node feature vector X, similarity S
    Output: Reconstructed node feature vector X'
    for each vector x = 1, 2, ..., i do
      for neighbor vector x' = 1, 2, ..., j do
        if X does not converge then
          Propagation feature:
          X = sum(S·x)/sum(S)
          Reconstruct known features:
          X' = X
        end if
      end for
    end for

    Let the feature matrix of the reconstructed patient similarity network $ \vartheta = (\nu, \epsilon) $ after feature completion be $ X $, denoting the feature vector of node i. First, the attention coefficients of each patient and its neighbor nodes are calculated, and then the features of the nodes are learned using the dynamic attention mechanism to obtain the vector representation of the patient nodes in the low-dimensional embedding space, and finally the fully connected layer is input for binary prediction of patient mortality or not. The specific operation process is as follows:

    Attention coefficient. For each patient node i, the weight vector of $ {h_{i, j}} $ its neighbor node is calculated, and the attention coefficient of $ {a_{i, j}} $ each neighbor node is obtained by the softmax function, which is calculated as follows:

    $ {a}_{i, j} = \frac{\mathrm{exp}(Leaky\mathrm{Re}LU(W·[{h}_{i, j}||{h}_{i}]))}{{\displaystyle {\sum }_{k\in {N}_{i}}\mathrm{exp}(Leaky\mathrm{Re}LU(W·[{h}_{i, k}||{h}_{i}]))}} $ (5)

    where $ {h_{i, j}}||{h_i} $ denotes the feature vector $ {h_i} $ of node i is stitched together with the feature vector $ {h_j} $ of neighbor node j. W is a learnable weight matrix and $ L{\text{eaky}}{Re} LU() $ is an activation function.

    Dynamic attention mechanism. The feature vector $ {h'_i} $ of node i is obtained by weighting and summing the feature vectors of neighboring nodes using the attention coefficients $ {a_{i, j}} $:

    $ {{h}^{\prime }}_{i} = {\displaystyle \sum\limits_{j\in {N}_{i}}{a}_{i, j}}·{h}_{j} $ (6)

    Feature representation learning. Using the feature vector $ h_i^{'} $ of node i as input, mapping it to a vector representation $ {z_i} $ in a low-dimensional embedding space through a fully connected layer:

    $ {z}_{i} = \mathrm{Re}LU({W}^{\prime }·{{h}^{\prime }}_{i})+b $ (7)

    where $ {Re} LU() $ is an activation function, $ W' $ is a linear transformation matrix and $ b $ is bias term.

    Prediction. The low-dimensional embedding vector $ {z_i} $ of nodes is used as the input of the fully connected layer, and the binary prediction result of whether the patient died is output.

    $ {y_i} = \sigma ({W^T}{h_i} + b) $ (8)

    where $ {y_i} $ denotes the prediction result of the node, $ {W^T} $ denotes the weight, $ b $ denotes the bias and $ \sigma $ denotes the softmax function.

    In summary, the process of ICU patient death risk prediction algorithm based on patient similarity dynamic graph attention network is as follows:

    Algorithm 2. Predicting the risk of mortality in ICU patients based on patient similarity dynamic graph attention network
    Input: feature matrix X of the feature-complemented patient similarity network $ \vartheta = (\nu, \epsilon) $
    Output: Prediction results for patient nodes $ {y_{\text{i}}} $
    for each node $ i \in v $ do
      Step 1: Calculate the weight vector $ {{\text{h}}_{{\text{i, j}}}} $ of node i's neighbor nodes and get the attention coefficient $ {a_{i, j}} $.
      Step 2: Use the attention coefficient to weighted sum the feature vectors $ {a_{i, j}} $ of neighbor nodes to obtain the feature vector $ {h'_i} $ of node $ i $.
      Step 3: Using the feature vector $ h_i^{'} $ of node $ i $ in the reconstructed patient similarity network as input, it is mapped to the vector representation $ {z_i} $ in the low-dimensional embedding space through a fully connected layer.
      Step 4: The low-dimensional embedding vector $ {z_i} $ of the node is used as the input of the fully connected layer to output the binary classification result $ {y_i} $ of whether the patient died.
    end for

    We used MIMIC-III v1.4, a large publicly available dataset containing approximately 60,000 patients treated in intensive care units (ICUs) from 2001 to 2012 [20], and screened data using the following criteria: 1) patients diagnosed with heart failure according to International Classification of Diseases codes (ICD-9 codes); 2) patients aged ≥ 18 years at the time of ICU admission; 3) repeat admissions or repeat ICU admissions of the same patient, and only data from their first ICU admission were included; 4) exclusion of patients with missing N-terminal brain natriuretic peptidogen (NT-proBNP) data. A total of 10,436 patients with a diagnosis of heart failure were queried, and 1255 adult patients were finally included in this study through screening.

    Based on previous studies [21,22,23], we mainly extracted four types of data: 1) demographic characteristics: age, gender, height and weight at admission; 2) comorbidities: including diabetes mellitus, hypertension, chronic kidney disease, chronic pulmonary obstruction, cardiac arrhythmia, iron-deficiency anemia, hyperlipidemia and atrial fibrillation; 3) vital signs: vital signs within the first 24 hours after admission to the ICU, including temperature, heart rate, blood pressure and respiratory rate; 4) laboratory tests: throughout the ICU stay Laboratory examination index values, including NTproBNP, BNP, troponin, serum creatinine, red blood cells, white blood cells, platelet counts and so on. The coding of patient characteristics is shown in Table 1.

    Table 1.  Information table of patient characteristics.
    Index Feature name Code
    1
    2
    3
    4
    ID
    hadm_id
    label
    age
    Numerical value
    Numerical value
    0 no dead, 1 dead
    Numerical value
    5 sex 1 male, 2 female
    6 BMI Numerical value
    7–15 comorbidity 0 no disease, 1 disease
    16–22
    23–51
    vital signs
    Laboratory tests
    Numerical value
    Numerical value

     | Show Table
    DownLoad: CSV

    After extracting all the features of the patients according to the above criteria, we further processed the data to make it directly usable for the model.

    Missing value processing. We first use the typical missing value processing method, KNN fill, to initially process the data for missing values. In order to improve the accuracy of the missing value filling, we initially processed the missing values based on the KNN method, we propose the feature completion method to deal with the missing values again by constructing the patient similarity network, and reconstruct the missing features by iteratively diffusing the known features in the graph, so as to further improve the accuracy of the prediction model. The specific method is stated in Section 3.2.

    Data standardization. We categorized the data into binary and numeric. 1) Binary features included sex and comorbidities, which were assigned values of 1 and 0, respectively, based on whether or not they were male and whether or not they suffered from that comorbidity. 2) Age, BMI, vital signs and laboratory tests were numeric features, and we normalized each feature using the min-max normalization method, which scales the feature values to between 0 and 1, to ensure that they have similar scales. This is shown in Eq (9).

    $ {X_{{\text{normalized}}}} = \frac{{X - {X_{{\text{min}}}}}}{{{X_{{\text{max}}}}{\text{ - }}{X_{{\text{min}}}}}} $ (9)

    After preprocessing the data as described above, we use the method in Section 3.1 to calculate patient similarity and construct a patient similarity network. The data size statistics of the final input model are shown in Table 2.

    Table 2.  Data set size statistics.
    Dataset Nodes Edges Features Training Validation Test
    MIMIC-III 1255 5336 48 1005 125 125

     | Show Table
    DownLoad: CSV

    In this paper, the experimental data set is divided into training set, verification set and test set according to 8:1:1. A dynamic graph attention network was implemented using the Python programming language and a PyTorch-based deep learning framework to predict the risk of patient mortality. To make the experimental results reproducible, this paper uses SEED = 42 as a random seed to ensure the same randomness for each experiment. The hyperparameters of the training model were n_epochs = 100, batch_size = 128 and lr = 0.001. CPU was used for model computation during the training process. To facilitate the observation of the model performance, the classification effect of the model on the test set is measured by three metrics: AUC, Accuracy and F1 score. All experiments were run in the same environment.

    The purpose of this study was to verify the effectiveness of the PS-DGAT model in predicting the risk of death in ICU patients. In order to verify the effect of the model, we compare it with traditional machine learning methods and use different graph convolution layers. The following comparison algorithms are mostly used.

    RandomForest (RF) [24]: Random forest is an integrated learning method that uses random feature selection and self-sampling techniques to improve model accuracy and generalization by combining multiple decision tree models for classification and regression.

    SVM [25]: It is a supervised learning algorithm for solving classification and regression problems by constructing optimal hyperplanes to establish decision boundaries between different classes of samples.

    Light GBM [26]: It is a gradient boosting decision tree algorithm for efficient model training and prediction through learning rate-based decision tree training and histogram algorithm acceleration.

    Decision trees [27]: are a typical classification method that first processes the data, generates readable rules and decision trees using inductive algorithms and then uses the decisions to analyze the new data. Essentially decision trees are a process of classifying data through a set of rules.

    Multilayer perceptron (MLP) [28]: is a feedforward artificial neural network model, which consists of input layer, hidden layer and output layer. The working principle is to add a hidden layer or a fully connected hidden layer between the output layer and the input layer, map multiple input data sets to a single output data set and convert the output of the hidden layer through the activation function.

    Graph Convolutional Neural Network (GCN) [29]: is a multilayer convolutional neural network where each convolutional layer processes only first-order neighborhood information, and by superimposing several convolutional layers, information transfer in multiple-order neighborhoods can be achieved.

    Graph attention network (GAT): Using attention mechanisms to learn relative importance between nodes and perform node classification or prediction tasks on graph data.

    GraphSAGE [30]: Learning node representations by aggregating neighboring features of nodes enables efficient node classification and embedding learning on large-scale graph data.

    Cluster-GCN [31]: Partitioning large-scale graph data into multiple subgraphs using a clustering algorithm and then performing a convolution operation on each subgraph.

    In order to verify the effectiveness of our proposed model on the mortality risk prediction task, we compared the overall performance of the model with the baseline models on the MIMIC-III dataset, using Accuracy, AUC and F1 score as evaluation indicators. The comparison results are shown in Table 3.

    Table 3.  Overall performance comparison of ours model and baseline models.
    Model evaluation metrics
    Accuracy (%) AUC (%) F1 score (%)
    RF 75.30 ± 025 70.53 ± 0.65 55.61 ± 0.29
    SVM 80.78 ± 1.22 72.77 ± 0.38 53.77 ± 0.64
    Light GBM 76.45 ± 0.30 69.45 ± 0.89 49.90 ± 0.44
    Decision Tree 79.20 ± 0.50 53.13 ± 0.21 45.86 ± 0.50
    MLP 87.46 ± 0.25 77.98 ± 0.53 60.50 ± 0.32
    GCN 87.35 ± 1.15 75.02 ± 0.25 62.28 ± 0.43
    GAT 87.96 ± 1.30 70.34 ± 0.67 63.30 ± 0.78
    GraphSAGE 87.20 ± 0.88 78.65 ± 0.38 52.86 ± 0.20
    Cluster-GCN 86.55 ± 0.34 75.48 ± 0.64 60.88 ± 0.56
    PS-DGAT 89.03 ± 0.98 80.92 ± 0.74 69.69 ± 0.23

     | Show Table
    DownLoad: CSV

    The above lists the performance of several models on the test set. From the above chart, it can be seen that compared with the other baseline models, the overall effect of the model in the prediction task is better than the traditional machine learning methods and GNN methods, and the Accuracy, AUC and F1 score values are higher than baseline model. The baseline models are categorized into traditional machine learning methods and graph neural network methods, and since the missing feature completion method proposed in this paper is based on graph models, the five machine learning models mentioned above, RF, SVM, Light GBM, Decision Tree and MLP, processed the missing values using only the traditional KNN filling method, and the other graph models use the feature completion method on top of the KNN filling to update the missing values. It can be found that the models that have used the missing feature completion method generally have higher overall performance. Moreover, in order to verify the effectiveness of the missing feature completion method for other graph models, we compare the graph baseline models that used feature completion with those that did not, and the comparison results are shown in Table 4.

    Table 4.  Comparison of feature completion for baseline models performance improvement.
    Model with feature completion without feature completion
    Accuracy (%) AUC (%) F1 score (%) Accuracy (%) AUC (%) F1 score (%)
    GCN 87.35 ± 1.15 75.02 ± 0.25 62.28 ± 0.43 85.65 ± 0.35 74.53 ± 0.64 60.17 ± 0.13
    GAT 87.96 ± 1.30 70.34 ± 0.67 63.30 ± 0.78 86.27 ± 0.23 69.44 ± 0.73 61.58 ± 0.11
    GraphSAGE 87.20 ± 0.88 78.65 ± 0.38 52.86 ± 0.20 85.63 ± 1.02 76.88 ± 0.26 50.49 ± 0.70
    Cluster-GCN 86.55 ± 0.34 75.48 ± 0.64 60.88 ± 0.56 85.11 ± 0.37 73.28 ± 0.74 58.91 ± 0.22
    PS-DGAT 89.03 ± 0.98 80.92 ± 0.74 69.69 ± 0.23 86.20 ± 0.68 79.17 ± 0.52 67.30 ± 0.66

     | Show Table
    DownLoad: CSV

    As shown in Table 4, compared to the traditional missing value processing method, our missing feature completion method has improved the performance of all other graph baseline models, and this result shows that the missing feature completion method is effective for the processing of missing features in graph models.

    The processing of missing values is a key step in data preprocessing. The final effect of many models depends largely on the effect of missing values processing. In order to select an effective missing value processing method and to verify the effectiveness of the feature completion method, the prediction results of the model without feature completion and with feature completion methods after the missing values are not processed, mean-completion, median-completion and KNN-completion are compared, and denoted as unprocessed, Mean, Median, KNN, Mean+FC, Median+FC and KNN+FC, respectively. Thus, we verify the necessity and superiority of the feature completion method, and the obtained results are shown in Figure 5.

    Figure 5.  Comparison of different missing value processing methods.

    Through the experiment results in Figure 5, for the MIMIC-III dataset, the use of KNN fill in this paper is better than other methods, and the feature completion algorithm to fill in the missing values has a significant improvement in the performance of the model and is superior to any traditional missing value processing methods. It shows that the effect of filling the missing features by propagating the information of neighboring nodes is superior to the use of missing value processing methods, such as mean and median. Thus, it verifies the effectiveness of the feature complementation method for the data in this paper.

    We reconstruct the missing features by iteratively propagating the known features of adjacent nodes on the graph, and make better use of the relationship between neighbor nodes to fill the missing eigenvalues, so as to obtain a complete feature matrix. Among them, the condition for the feature iteration to stop is that the iteration stops when the feature vector no longer changes or changes slightly or reaches the maximum number of iterations (40 times). We explain the difference in the prediction results of the feature completion method under different iterations, as shown in Figure 6.

    Figure 6.  Effect of iteration times on experimental results.

    Figure 6 shows the difference in the prediction results of the feature completion method under different iterations. We observe that with fewer iterations, the prediction results may not make full use of the relationship between neighbor nodes for feature completion, so there may be a large error. As the number of iterations increases, the prediction results gradually converge and stabilize, and the error gradually decreases. However, when a certain number of iterations is reached, further increasing the number of iterations has no obvious effect on the improvement of the results, and the change of the feature vector is very small. Therefore, according to the experimental results and the results of convergence analysis, we chose the maximum number of iterations as 40 times.

    Because setting different thresholds will produce different network structures, which will affect the classification results of the model, this paper analyzes the effect of setting the similarity threshold between 0 and 1 on the accuracy of the model. The experimental results are shown in Figure 7. From the figure, it can be seen that the threshold value has a significant effect on the experimental results. As the threshold value gradually increases from 0 to 0.5, the classification accuracy of the model also keeps improving. Because the model is classified according to the similarity of nodes, the higher the similarity, the more accurate the classification result. So, when the threshold value increases from 0, more and more less similar nodes are filtered out, and the effective impact of the embedding representation of the node similarity learning nodes based on local information will be smaller and smaller until all similar nodes are filtered out and the experimental results return to the initial state. The horizontal line in the figure indicates the classification accuracy of this paper's model when other parameters are the same. It can be found that the model accuracy is highest when the similarity threshold is set at 0.5, so the threshold is set to 0.5 in this paper.

    Figure 7.  Effect of similarity threshold on accuracy.

    In order to better understand and explain the effect of the PS-DGAT model, we perform dimensionality reduction visualization analysis on the PS-DGAT model. Using t-SNE [32] dimensionality reduction technology, the patient feature space is mapped to a two-dimensional plane. The relative position and similarity relationship between patients are displayed visually, and the distribution and clustering patterns of patients in the two-dimensional space can be observed.

    For Figure 8(a) initial feature distribution, it can be observed that the patient's feature vectors show a certain aggregation trend. This means that in the initial state, the patient's characteristics are relatively consistent in the numerical range, or have similar characteristic patterns. This may be due to the influence of data preprocessing, in which the features are normalized, scaled or outliers are removed. In addition, it is also possible that the patient data has similar pathological state or biological characteristics, resulting in similar feature vectors in the initial state. It should be noted that the aggregation of the initial feature distribution does not necessarily reflect the real differences and similarities between patients. Therefore, we explore the similarity between patients through further dimensionality reduction and visual analysis, and more accurately capture the feature patterns and correlations of patient data.

    Figure 8.  Visualization results of PS-DGAT model. (a) Initial features; (b) Features reconstructed after feature completion; (c) Features after node embedding.

    For Figure 8(b) feature distribution reconstructed after feature completion, it can be observed that the patient nodes show a more dispersed distribution in the dimensionality reduction space, and the distribution of the nodes is relatively uniform. This shows that through the iterative propagation and similarity calculation of neighbor node features, we successfully recover the feature differences between patients and accurately capture the change patterns and correlations of patient data. In the process of feature completion, the feature information of neighbor nodes is effectively transmitted and utilized, so that the patient nodes can be better separated and expanded in the dimensionality reduction space.

    For Figure 8(c) feature distribution of nodes after embedding, it can be observed that the distribution of nodes shows the same color aggregation phenomenon, which means that we have successfully completed the classification task of nodes. Nodes with the same color clustering represent that they are classified as the same category or cluster, with similar feature patterns and attributes. This shows that our model effectively captures the similarities and differences between different patients when learning node embedding, so that similar nodes are close to each other in the embedding space, and different nodes are dispersed. Through this node aggregation phenomenon, we can clearly observe the distribution of patient data in the feature space, and further data analysis and interpretation can be performed based on these aggregation results.

    In this paper, we propose a patient similarity-based dynamic graph attention network (PS-DGAT) for predicting the risk of patient mortality in ICU by combining node similarity with graph attention network, and first constructing a patient similarity network by calculating the similarity of patient clinical data to transform the structured data into a graph structure. Then, using similarity, missing features were entered and the patient similarity network was updated based on the data of adjacent patients in the network. Finally, the PS-DGAT model with adjusted order of weighted vector operations in GAT was used to predict the mortality of ICU patients.

    In this paper, experimental validation is performed on the MIMIC-III dataset, and the experimental results show that PS-DGAT has better prediction effect than the traditional classification methods, such as GAT, which proves the effectiveness of the model in this paper. The model in this paper also has shortcomings: only structured data are used for clinical data, which may have low prediction accuracy, and unstructured text data should be added in future studies to improve the model performance. In addition, when filling in the missing data in this paper, it is necessary to reconstruct the patient similarity network after reconstructing the features, which increases the complexity, and better methods can be proposed in subsequent studies to improve the prediction accuracy while reducing the computational complexity and workload.

    The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.

    We would like to thank the anonymous reviewers for their invaluable comments, whose valuable suggestions and comments contributed significantly to this study.

    The authors declare there is no conflict of interest.

    [1] Ministry of Agriculture and Food Security, Final results of the agricultural campaign and the food and nutrition situation, Departmental Report, 2014. Available from: http://cns.bf/IMG/pdf/rapport_general_des_resultats_definitifs_2012_2013.pdf.
    [2] Ministère de la santé, Direction de la Nutrition, Enquête nutritionnelle nationale 2016. Rapport du Ministère, 2016. Available from: https://www.humanitarianresponse.info/system/files /documents/files/smart_2016.pdf.
    [3] FAO, L'état de l'insécurité alimentaire dans le monde 2013. Les multiples dimensions de la sécurité alimentaire, Rome, FAO ISBN 978-92-5-207916-3 (version imprimée), E-ISBN 978-92-5-207917-0, 2013. Available from: http://www.fao.org/docrep/019/i3434f/i3434f.pdf.
    [4] Trèche S (1999) Aliments de complément: caractéristiques nutritionnelles et hygiéniques, production, utilisation.
    [5] Tou ElH, Mouquet-Rivier C, Picq C, et al. (2007) Improving the nutritional quality of ben-saalga, a traditional fermented millet-based gruel, by co-fermenting millet with groundnut and modifying the processing method. Food Sci Technol 40: 1561–1569.
    [6] Kyelem CG, Bougouma A, Thiombiano RS, et al. (2011) Cholera outbreak in Burkina Faso in 2005: epidemiological and diagnostic aspects. Pan Afr Med J 8: 1.
    [7] Akaki KD, Sadat AW, Loiseau G, et al. (2008) Étude du comportement des souches de Bacillus cereus atcc 9139 et d'Escherichia coli atcc 25922 par la méthode des challenge-tests lors de la confection de bouillies à base de pate de mil fermentée en provenance de Ouagadougou (Burkina-Faso). Rev Ivoir Sci Technol 11: 103–117.
    [8] OMS, Maladies d'origine alimentaire: près d'un tiers des décès surviennent chez les enfants de moins de 5 ans. Rapport sur l'impact des aliments contaminés sur la santé et le bien-être, 2015. Available from: http://www.who.int/mediacentre/news/releases/2015/foodborne-disease-estimates/fr/.
    [9] Barro N, Sangaré L, Tahita MC, et al. (2005) Les principaux agents du péril identifiés dans les aliments de rue et ceux des cantines et leur prévalence en milieu hospitalier. Maîtrise des Procédés en vue d'améliorer la qualité et la sécurité des aliments, utilisation des OGM, analyse des risques en agroalimentaire, Séminaire régional du réseau GP3A, Ouagadougou, 8–11.
    [10] Bonkoungou IJ, Damanka S, Sanou I, et al. (2011) Genotype diversity of group A rotavirus strains in children with acute diarrhea in urban Burkina Faso, 2008–2010. J Med Virol 83: 1485–1490. doi: 10.1002/jmv.22137
    [11] Bessimbaye N, Tidjani A, Gamougame K, et al. (2013) Gastro-enterites en milieux des réfugiés au Tchad. IJBCS 7: 468–478.
    [12] Ezekiel CN, Udomb IE, Frisvad JC, et al. (2014) Assessment of aflatoxigenic Aspergillus and other fungi in millet and sesame from Plateau State, Nigeria. Mycology 1: 16–22.
    [13] Afolabi CG, Ezekiel CN, Kehinde IA, et al. (2015) Contamination of Groundnut in South-Western Nigeria by Aflatoxigenic Fungi and Aflatoxins in Relation to Processing. J Phytopathol 163: 279–286. doi: 10.1111/jph.12317
    [14] Mouquet C, Bruyeron O, Trèche S (1998) Caractéristiques d'une Bonne Farine Infantile. Bulletin du Réseau TPA 15: 8–11.
    [15] Au Togo E (1998) Dossier: Les farines infantiles. Bulletin du réseau TPA 15: 211–213.
    [16] Ministry of health, Microbiological criteria applicable to foodstuffs: Guidelines for interpretation, edition November, 2015. Available from: http://www.securite-alimentaire.public.lu/professionnel/denrees_alimentaires/qualite_aliments/recueil_criteres_microbiologiques/recueil_criteres_microbiologiques.pdf.
    [17] Tou ElH, Mouquet-Rivier C, Rochette I, et al. (2007) Effect of different process combinations on the fermentation kinetics, microflora and energy density of ben-saalga, a fermented gruel from Burkina Faso. Food Chem 100: 935–943. doi: 10.1016/j.foodchem.2005.11.007
    [18] FAO/WHO, Enterobacter sakazakii (Cronobacter spp.) in powdered follow-up formulae. Microbiological risk assessment series 15, Meeting report, 2008. Available from: http://www.fao.org/fileadmin/templates/agns/pdf/jemra/mra15_sakazaki.pdf.
    [19] Kayalto B, Zongo C, Compaoré RW, et al. (2013) Study of the nutritional value and hygienic quality of local infant flours from Chad, with the aim of their use for improved infant flours preparation. Food Nutr Sci 4: 59–68.
    [20] Matug SM, Aidoo KE, Elgerbi AM (2015) Microbiological examination of infant food and feed formula. Emerg Life Sci Res 1: 46–51.
    [21] Norberg S, Stanton C, Ross RP, et al. (2012) Cronobacter spp. in powdered infant formula. J Food Protect 75: 607–620.
    [22] Food safety (Authority of IRELAND) (2011) Cronobacter spp. (Enterobacter sakazakii), In: Microbial fact sheet series.
    [23] Agence de la sante publique du Canada, Klebsiella spp.: Fiche technique sante-securite, agents pathogenes, 2011. Available from: http://www.phac-aspc.gc.ca/lab-bio/res/psds-ftss/klebsiella-fra.php.
    [24] Anses (Angence national de sécurité sanitaire, aliment, environnement, travail), Cronobacter spp. Enterobacter sakazakii Famille des Enterobacteriaceae Bactérie. Fiche de description de danger biologique transmissible par les aliments/Cronobacter spp., 2011. Available from: https://www.anses.fr/fr/system/files/MIC2000sa0003Fi.pdf.
    [25] Alborch L, Bragulat MR, Castellá G, et al. (2012) Mycobiota and mycotoxins contamination of maize flours and popcorn kernels for human consumption commercialized in Spain. Food Microbiol 32: 97–103. doi: 10.1016/j.fm.2012.04.014
    [26] Waré LY, Durand N, Nikiema PA, et al. (2017) Occurrence of mycotoxins in commercial infant formulas locally produced in Ouagadougou (Burkina Faso). Food Control 73: 518–523. doi: 10.1016/j.foodcont.2016.08.047
    [27] Abarca ML, Bragulat MR, Castella G, et al. (1994) Ochratoxin A production by strains of Aspergillus niger Var niger. Appl Environ Microb 60: 2650–2652.
    [28] Heenan CN, Shaw KJ, Pitt JI (1998) Ochratoxin A production by Aspergillus carbonarius and Aspergillus niger isolates and detection using coconut cream agar. J Food Mycol 1: 67–72.
    [29] Taniwaki MH, Pitt JI, Teixeira AA, et al. (2003) The source of ochratoxin A in Brazilian coffee and its formation in relation to processing methods. Int J Food Microbiol 82: 173–179. doi: 10.1016/S0168-1605(02)00310-0
    [30] Jayaramachandran R, Ghadevaru S, Veerapandian S (2013) Survey of market samples of food grains and grain flour for Aflatoxin B1 contamination. IJCMAS 2: 184–188.
    [31] Egal S, Hounsa A, Gong YY, et al. (2005) Dietary exposure to aflatoxin from maize and groundnut in youngchildren from Benin and Togo, West Africa. Int J Food Microbiol 104: 215–224. doi: 10.1016/j.ijfoodmicro.2005.03.004
    [32] Raiola A, Tenore GC, Manyes L, et al. (2015) Risk analysis of main mycotoxins occurring in food for children: an overview. Food Chem Toxicol 84: 169–180. doi: 10.1016/j.fct.2015.08.023
    [33] USAID, Aflatoxine: une synthese de la recherche en sante, agriculture et commerce, rapport final version francaise Mission Regionale de l'Afrique de l'Est Nairobi, Kenya, 2012. Available from: http://www.aflatoxinpartnership.org/uploads/Aflatoxin%20Desk%20Study%20Final%20 Report%202012%20French%20Version.pdf.
  • This article has been cited by:

    1. Niki Baccile, Florence Babonneau, Ibrahim M. Banat, Katarzyna Ciesielska, Anne-Sophie Cuvier, Bart Devreese, Bernd Everaert, Helen Lydon, Roger Marchant, Christopher A. Mitchell, Sophie Roelants, Lasse Six, Elke Theeuwes, Georgios Tsatsos, Georgia Eleni Tsotsou, Brecht Vanlerberghe, Inge N. A. Van Bogaert, Wim Soetaert, Development of a Cradle-to-Grave Approach for Acetylated Acidic Sophorolipid Biosurfactants, 2017, 5, 2168-0485, 1186, 10.1021/acssuschemeng.6b02570
    2. Cristiana Pontes, Marta Alves, Catarina Santos, Maria H. Ribeiro, Lídia Gonçalves, Ana F. Bettencourt, Isabel A.C. Ribeiro, Can Sophorolipids prevent biofilm formation on silicone catheter tubes?, 2016, 513, 03785173, 697, 10.1016/j.ijpharm.2016.09.074
    3. Sib Sankar Giri, Hyoun Joong Kim, Sang Guen Kim, Sang Wha Kim, Jun Kwon, Sung Bin Lee, Se Chang Park, Immunomodulatory Role of Microbial Surfactants, with Special Emphasis on Fish, 2020, 21, 1422-0067, 7004, 10.3390/ijms21197004
    4. Ana Letícia Silva Coelho, Paulo Emilio Feuser, Bruno Augusto Mattar Carciofi, Débora de Oliveira, Cristiano José de Andrade, Biological activity of mannosylerythritol lipids on the mammalian cells, 2020, 104, 0175-7598, 8595, 10.1007/s00253-020-10857-9
    5. Chiara Ceresa, Maurizio Rinaldi, Letizia Fracchia, Synergistic activity of antifungal drugs and lipopeptide AC7 against Candida albicans biofilm on silicone, 2017, 4, 2375-1495, 318, 10.3934/bioeng.2017.2.318
    6. Drifa Yalaoui-Guellal, Samira Fella-Temzi, Salima Djafri-Dib, Fatiha Brahmi, Ibrahim M. Banat, Khodir Madani, Biodegradation potential of crude petroleum by hydrocarbonoclastic bacteria isolated from Soummam wadi sediment and chemical-biological proprieties of their biosurfactants, 2020, 184, 09204105, 106554, 10.1016/j.petrol.2019.106554
    7. Lakshmi Tripathi, Matthew S. Twigg, Aikaterini Zompra, Karina Salek, Victor U. Irorere, Tony Gutierrez, Georgios A. Spyroulias, Roger Marchant, Ibrahim M. Banat, Biosynthesis of rhamnolipid by a Marinobacter species expands the paradigm of biosurfactant synthesis to a new genus of the marine microflora, 2019, 18, 1475-2859, 10.1186/s12934-019-1216-8
    8. Breedge Callaghan, Helen Lydon, Sophie L. K. W. Roelants, Inge N. A. Van Bogaert, Roger Marchant, Ibrahim M. Banat, Christopher A. Mitchell, Chunming Liu, Lactonic Sophorolipids Increase Tumor Burden in Apcmin+/- Mice, 2016, 11, 1932-6203, e0156845, 10.1371/journal.pone.0156845
    9. Ibrahim M. Banat, Quentin Carboué, Gerardo Saucedo-Castañeda, José de Jesús Cázares-Marinero, Biosurfactants: The green generation of speciality chemicals and potential production using Solid-State fermentation (SSF) technology, 2021, 320, 09608524, 124222, 10.1016/j.biortech.2020.124222
    10. Sonam Gupta, Navdeep Raghuwanshi, Ritu Varshney, I.M. Banat, Amit Kumar Srivastava, Parul A. Pruthi, Vikas Pruthi, Accelerated in vivo wound healing evaluation of microbial glycolipid containing ointment as a transdermal substitute, 2017, 94, 07533322, 1186, 10.1016/j.biopha.2017.08.010
    11. T. P. Pirog, SYNERGISTIC ACTION OF ESSENTIAL OILS WITH THE BIOCIDES ON MICROORGANISMS, 2019, 12, 24107751, 5, 10.15407/biotech12.04.005
    12. Rachhpal S. Kahlon, 2016, Chapter 8, 978-3-319-31197-5, 281, 10.1007/978-3-319-31198-2_8
    13. Yuan-Seng Wu, Siew-Ching Ngai, Bey-Hing Goh, Kok-Gan Chan, Learn-Han Lee, Lay-Hong Chuah, Anticancer Activities of Surfactin and Potential Application of Nanotechnology Assisted Surfactin Delivery, 2017, 8, 1663-9812, 10.3389/fphar.2017.00761
    14. B. Jasim, S. Sreelakshmi, Jyothis Mathew, E. K. Radhakrishnan, Identification of endophytic Bacillus mojavensis with highly specialized broad spectrum antibacterial activity, 2016, 6, 2190-572X, 10.1007/s13205-016-0508-5
    15. C. Ceresa, L. Fracchia, M. Williams, I.M. Banat, M.A. Díaz De Rienzo, The effect of sophorolipids against microbial biofilms on medical-grade silicone, 2020, 309, 01681656, 34, 10.1016/j.jbiotec.2019.12.019
    16. T.P. Pirog, L.V. Nikituk, T.A. Shevchuk, Synergism of Antimicrobial Activity of Nocardia vaccinii IMV B-7405 Surfactants and Antibiotics, 2017, 79, 10280987, 30, 10.15407/microbiolj79.04.030
    17. Chiara Ceresa, Francesco Tessarolo, Devid Maniglio, Erica Tambone, Irene Carmagnola, Emanuele Fedeli, Iole Caola, Giandomenico Nollo, Valeria Chiono, Gianna Allegrone, Maurizio Rinaldi, Letizia Fracchia, Medical-Grade Silicone Coated with Rhamnolipid R89 Is Effective against Staphylococcus spp. Biofilms, 2019, 24, 1420-3049, 3843, 10.3390/molecules24213843
    18. Victória Akemi I. Silveira, Erick K Nishio, Christiane A.U.Q. Freitas, Ismael R. Amador, Renata K.T. Kobayashi, Talita Caretta, Fernando Macedo, Maria Antonia P.C. Celligoi, Production and antimicrobial activity of sophorolipid against Clostridium perfringens and Campylobacter jejuni and their additive interaction with lactic acid, 2019, 21, 18788181, 101287, 10.1016/j.bcab.2019.101287
    19. Bac V. G. Nguyen, Toshiki Nagakubo, Masanori Toyofuku, Nobuhiko Nomura, Andrew S. Utada, Synergy between Sophorolipid Biosurfactant and SDS Increases the Efficiency of P. aeruginosa Biofilm Disruption, 2020, 36, 0743-7463, 6411, 10.1021/acs.langmuir.0c00643
    20. Mohamed Elshikh, Roger Marchant, Ibrahim M. Banat, Geertje van Keulen, Biosurfactants: promising bioactive molecules for oral-related health applications, 2016, 363, 1574-6968, fnw213, 10.1093/femsle/fnw213
    21. T.P. Pirog, O.I. Paliichuk, G.O. Iutynska, T.A. Shevchuk, Prospects of Using Microbial Surfactants in Plant Growing, 2018, 80, 10280987, 115, 10.15407/microbiolj80.03.115
    22. Italo José Batista Durval, Ivison Amaro da Silva, Leonie Asfora Sarubbo, 2021, Chapter 1, 978-981-15-6606-6, 1, 10.1007/978-981-15-6607-3_1
    23. P.J. Naughton, R. Marchant, V. Naughton, I.M. Banat, Microbial biosurfactants: current trends and applications in agricultural and biomedical industries, 2019, 127, 1364-5072, 12, 10.1111/jam.14243
    24. Ana Letícia Silva Coelho, Paulo Emílio Feuser, Bruno Augusto Mattar Carciofi, Cristiano José de Andrade, Débora de Oliveira, Mannosylerythritol lipids: antimicrobial and biomedical properties, 2020, 104, 0175-7598, 2297, 10.1007/s00253-020-10354-z
    25. Chiara Ceresa, Maurizio Rinaldi, Francesco Tessarolo, Devid Maniglio, Emanuele Fedeli, Erica Tambone, Patrizio Caciagli, Ibrahim M. Banat, Mayri Alessandra Diaz De Rienzo, Letizia Fracchia, Inhibitory Effects of Lipopeptides and Glycolipids on C. albicans–Staphylococcus spp. Dual-Species Biofilms, 2021, 11, 1664-302X, 10.3389/fmicb.2020.545654
    26. Matthew L. Smith, Stefano Gandolfi, Philippa M. Coshall, Pattanathu K. S. M. Rahman, Biosurfactants: A Covid-19 Perspective, 2020, 11, 1664-302X, 10.3389/fmicb.2020.01341
    27. Constantina Eleni Drakontis, Samiul Amin, Biosurfactants: Formulations, properties, and applications, 2020, 48, 13590294, 77, 10.1016/j.cocis.2020.03.013
    28. Surekha K. Satpute, Gauri R. Kulkarni, Arun G. Banpurkar, Ibrahim M. Banat, Nishigandha S. Mone, Rajendra H. Patil, Swaranjit Singh Cameotra, Biosurfactant/s from Lactobacilli species: Properties, challenges and potential biomedical applications, 2016, 56, 0233111X, 1140, 10.1002/jobm.201600143
    29. Danyelle Santos, Raquel Rufino, Juliana Luna, Valdemir Santos, Leonie Sarubbo, Biosurfactants: Multifunctional Biomolecules of the 21st Century, 2016, 17, 1422-0067, 401, 10.3390/ijms17030401
    30. Benjamin Trudgeon, Markus Dieser, Narayanaganesh Balasubramanian, Mitch Messmer, Christine M. Foreman, Low-Temperature Biosurfactants from Polar Microbes, 2020, 8, 2076-2607, 1183, 10.3390/microorganisms8081183
    31. Zulfiqar Ali Raza, Zafar M. Khalid, Naseer Ahmad, Bushra Tehseen, Statistical Optimisation of Rhamnolipid Production using a Pseudomonas putida Strain Cultivated on Renewable Carbon Sources of Waste Vegetable Oils, 2020, 57, 0932-3414, 13, 10.3139/113.110664
    32. Farazul Haque, Mohd Sajjad Ahmad Khan, Naif AlQurashi, ROS-Mediated Necrosis by Glycolipid Biosurfactants on Lung, Breast, and Skin Melanoma Cells, 2021, 11, 2234-943X, 10.3389/fonc.2021.622470
    33. Amedea Perfumo, Michelle Rudden, Roger Marchant, Ibrahim M. Banat, 2018, Chapter 35, 978-3-319-50540-4, 75, 10.1007/978-3-319-50542-8_35
    34. Milena G. Rikalovic, Natasa S. Avramovic, Ivanka M. Karadzic, 2017, Chapter 6, 978-953-51-3291-2, 10.5772/66687
    35. A. Pele Milagre, Montero-Rodriguez Dayana, Rubio-Ribeaux Daylin, F. Souza Adriana, A. C. Luna Marcos, F. Santiago Michele, F. S. Andrade Rosileide, A. Lima e Silva Thayse, L. C. M. A. Santiago André, M. Campos-Takaki Galba, Development and improved selected markers to biosurfactant and bioemulsifier production by Rhizopus strains isolated from Caatinga soil, 2018, 17, 1684-5315, 150, 10.5897/AJB2017.16230
    36. Daniel Chakhalian, Robert B. Shultz, Catherine E. Miles, Joachim Kohn, Opportunities for biomaterials to address the challenges of COVID ‐19 , 2020, 108, 1549-3296, 1974, 10.1002/jbm.a.37059
    37. Basavaprabhu H. Nataraj, Syed Azmal Ali, Pradip V. Behare, Hariom Yadav, Postbiotics-parabiotics: the new horizons in microbial biotherapy and functional foods, 2020, 19, 1475-2859, 10.1186/s12934-020-01426-w
    38. Usharani Rathinam Krishnaswamy, Lakshmanaperumalsamy P, Achlesh D, 2021, Chapter 4, 978-981-15-8998-0, 81, 10.1007/978-981-15-8999-7_4
    39. Shanying He, Yaoqi Ni, Li Lu, Qiwei Chai, Tao Yu, Zhiqiang Shen, Chunping Yang, Simultaneous degradation of n-hexane and production of biosurfactants by Pseudomonas sp. strain NEE2 isolated from oil-contaminated soils, 2020, 242, 00456535, 125237, 10.1016/j.chemosphere.2019.125237
    40. T. P. Pirog, THE PROPERTIES OF SURFACTANTS SYNTHESIZED BY Acinetobacter calcoaceticus ІMV В-7241 ON REFINED AND WASTE SUNFLOWER OIL, 2018, 11, 24107751, 82, 10.15407/biotech11.06.082
    41. M. Elshikh, I. Moya-Ramírez, H. Moens, S. Roelants, W. Soetaert, R. Marchant, I.M. Banat, Rhamnolipids and lactonic sophorolipids: natural antimicrobial surfactants for oral hygiene, 2017, 123, 13645072, 1111, 10.1111/jam.13550
    42. Chiara Ceresa, Francesco Tessarolo, Devid Maniglio, Iole Caola, Giandomenico Nollo, Maurizio Rinaldi, Letizia Fracchia, Inhibition of Candida albicans biofilm by lipopeptide AC7 coated medical-grade silicone in combination with farnesol, 2018, 5, 2375-1495, 192, 10.3934/bioeng.2018.3.192
    43. Sweeta Akbari, Nour Hamid Abdurahman, Rosli Mohd Yunus, Fahim Fayaz, Oluwaseun Ruth Alara, Biosurfactants—a new frontier for social and environmental safety: a mini review, 2018, 2, 24520721, 81, 10.1016/j.biori.2018.09.001
    44. Surekha K. Satpute, Nishigandha S. Mone, Parijat Das, Ibrahim M. Banat, Arun G. Banpurkar, Inhibition of pathogenic bacterial biofilms on PDMS based implants by L. acidophilus derived biosurfactant, 2019, 19, 1471-2180, 10.1186/s12866-019-1412-z
    45. Thaina A. Lima, Augusto Etchegaray, M. Teresa Machini, Design, synthesis and valued properties of surfactin oversimplified analogues, 2020, 52, 0939-4451, 25, 10.1007/s00726-019-02806-2
    46. Geeta Rawat, Anupam Dhasmana, Vivek Kumar, Biosurfactants: the next generation biomolecules for diverse applications, 2020, 3, 2523-8922, 353, 10.1007/s42398-020-00128-8
    47. Louiza Derguine-Mecheri, Salima Kebbouche-Gana, Djamel Djenane, Biosurfactant production from newly isolated Rhodotorula sp.YBR and its great potential in enhanced removal of hydrocarbons from contaminated soils, 2021, 37, 0959-3993, 10.1007/s11274-020-02983-3
    48. T.P. Pirog, T.A. Shevchuk, N.M. Petrenko, O.I. Paliichuk, G.O. Iutynska, Influence of Cultivation Conditions of Rhodococcus erythropolis IMV Ac-5017 on the Properties of Synthesized Surfactants, 2018, 80, 10280987, 13, 10.15407/microbiolj80.04.013
    49. Mohamed Elshikh, Scott Funston, Alif Chebbi, Syed Ahmed, Roger Marchant, Ibrahim M. Banat, Rhamnolipids from non-pathogenic Burkholderia thailandensis E264: Physicochemical characterization, antimicrobial and antibiofilm efficacy against oral hygiene related pathogens, 2017, 36, 18716784, 26, 10.1016/j.nbt.2016.12.009
    50. Chiara Ceresa, Maurizio Rinaldi, Valeria Chiono, Irene Carmagnola, Gianna Allegrone, Letizia Fracchia, Lipopeptides from Bacillus subtilis AC7 inhibit adhesion and biofilm formation of Candida albicans on silicone, 2016, 109, 0003-6072, 1375, 10.1007/s10482-016-0736-z
    51. Scott J. Funston, Konstantina Tsaousi, Thomas J. Smyth, Matthew S. Twigg, Roger Marchant, Ibrahim M. Banat, Enhanced rhamnolipid production in Burkholderia thailandensis transposon knockout strains deficient in polyhydroxyalkanoate (PHA) synthesis, 2017, 101, 0175-7598, 8443, 10.1007/s00253-017-8540-x
    52. Amedea Perfumo, Michelle Rudden, Roger Marchant, Ibrahim M. Banat, 2017, Chapter 35-1, 978-3-319-20796-4, 1, 10.1007/978-3-319-20796-4_35-1
    53. Chiara Ceresa, Letizia Fracchia, Emanuele Fedeli, Chiara Porta, Ibrahim M. Banat, Recent Advances in Biomedical, Therapeutic and Pharmaceutical Applications of Microbial Surfactants, 2021, 13, 1999-4923, 466, 10.3390/pharmaceutics13040466
    54. Kenia Barrantes, Juan José Araya, Luz Chacón, Rolando Procupez‐Schtirbu, Fernanda Lugo, Gabriel Ibarra, Víctor H. Soto, 2021, 9781119671022, 245, 10.1002/9781119671022.ch11
    55. Maria H. Ribeiro, Eva Fahr, Sara Lopes, 2022, 9781119769576, 1, 10.1002/9781119769620.ch1
    56. Marcos André Moura Dias, Marcia Nitschke, Bacterial-derived surfactants: an update on general aspects and forthcoming applications, 2023, 54, 1517-8382, 103, 10.1007/s42770-023-00905-7
    57. JINDARAT EKPRASERT, SASIWIMON KANAKAI, SULADDA YOSPRASONG, Improved Biosurfactant Production by Enterobacter cloacae B14, Stability Studies, and its Antimicrobial Activity, 2020, 69, 2544-4646, 273, 10.33073/pjm-2020-030
    58. Neelam Mishra, Kavita Rana, Siva Deepthi Seelam, Rakesh Kumar, Vijyendra Pandey, Bharathi P. Salimath, Dayanand Agsar, Characterization and Cytotoxicity of Pseudomonas Mediated Rhamnolipids Against Breast Cancer MDA-MB-231 Cell Line, 2021, 9, 2296-4185, 10.3389/fbioe.2021.761266
    59. Zahraa Amer Hashim, Jean‐Yves Maillard, Melanie Jay Wilson, Rachel Jane Waddington, Determining the potential use of biosurfactants in preventing endodontic infections, 2022, 130, 0909-8836, 10.1111/eos.12900
    60. Monohar Hossain Mondal, Wasefa Begum, Ajaya Bhattarai, Dileep Kumar, Bula Singh, Bidyut Saha, 2023, 9780128242834, 57, 10.1016/B978-0-12-824283-4.00002-2
    61. Moumita Biswas (Sarkar), Abhijit Sarkar, 2022, 9780128229217, 303, 10.1016/B978-0-12-822921-7.00016-7
    62. Simms A. Adu, Matthew S. Twigg, Patrick J. Naughton, Roger Marchant, Ibrahim M. Banat, Characterisation of cytotoxicity and immunomodulatory effects of glycolipid biosurfactants on human keratinocytes, 2023, 107, 0175-7598, 137, 10.1007/s00253-022-12302-5
    63. Anna Parus, Tomasz Ciesielski, Marta Woźniak-Karczewska, Mariusz Ślachciński, Mikołaj Owsianiak, Łukasz Ławniczak, Andreas P. Loibner, Hermann J. Heipieper, Łukasz Chrzanowski, Basic principles for biosurfactant-assisted (bio)remediation of soils contaminated by heavy metals and petroleum hydrocarbons – A critical evaluation of the performance of rhamnolipids, 2023, 443, 03043894, 130171, 10.1016/j.jhazmat.2022.130171
    64. Victor E. Vera-Santander, Ricardo H. Hernández-Figueroa, María T. Jiménez-Munguía, Emma Mani-López, Aurelio López-Malo, Health Benefits of Consuming Foods with Bacterial Probiotics, Postbiotics, and Their Metabolites: A Review, 2023, 28, 1420-3049, 1230, 10.3390/molecules28031230
    65. Aarati Panchabhai, 2021, 9780128233801, 367, 10.1016/B978-0-12-823380-1.00010-1
    66. Shristi Handa, Yadu Aggarwal, Sanjeev Puri, Mary Chatterjee, Pharmaceutical prospects of biosurfactants produced from fungal species, 2022, 62, 0233-111X, 1307, 10.1002/jobm.202200327
    67. Bhoomika M. Karamchandani, Ameya A. Pawar, Sujit S. Pawar, Sahil Syed, Nishigandha S. Mone, Sunil G. Dalvi, Pattanathu K. S. M. Rahman, Ibrahim M. Banat, Surekha K. Satpute, Biosurfactants’ multifarious functional potential for sustainable agricultural practices, 2022, 10, 2296-4185, 10.3389/fbioe.2022.1047279
    68. Amrita Kumari Panda, Seeta Dewali, Rashi Miglani, Rojita Mishra, Surajit De Mandal, Satpal Singh Bisht, 2023, 9780323901130, 147, 10.1016/B978-0-323-90113-0.00006-7
    69. Sweeta Akbari, Nour Hamid Abdurahman, Viacheslau Kudrashou, 2023, Chapter 7, 978-3-031-21681-7, 137, 10.1007/978-3-031-21682-4_7
    70. Celia Abolnik, Christine Strydom, Debbie Landman, Reneé Pieterse, Identification of bacteria in the tracheal swabs of farmed ostriches and their effect on the viability of influenza A virus, 2021, 33, 1040-6387, 1089, 10.1177/10406387211034483
    71. Sherly Antony, T.U. Sukumaran, Prasanth Rathinam, Reshmy R., Parameswaran Binod, Ashok Pandey, Raveendran Sindhu, 2022, 9780323851466, 439, 10.1016/B978-0-323-85146-6.00017-6
    72. Marcia Nitschke, Crisiane Aparecida Marangon, Microbial surfactants in nanotechnology: recent trends and applications, 2022, 42, 0738-8551, 294, 10.1080/07388551.2021.1933890
    73. Charles Oluwaseun Adetunji, Olugbemi Tope Olaniyan, Osikemekha Anthony Anani, Abel Inobeme, Awotunde Oluwasegun Samson, Julius Kola Oloke, Wadzani Dauda Palnam, Sana Ali, 2022, 9780323851466, 175, 10.1016/B978-0-323-85146-6.00014-0
    74. N.V. Krishna Prasad, R. Jayakumar, P. Wagdevi, S. Ramesh, 2021, 9780128226964, 21, 10.1016/B978-0-12-822696-4.00021-8
    75. Soni Kumari Singh, Ashish Sachan, 2022, 9780128234433, 419, 10.1016/B978-0-12-823443-3.00007-7
    76. Manoj Kumar Sarangi, Sasmita Padhi, L.D. Patel, Goutam Rath, Sitansu Sekhar Nanda, Dong Kee Yi, Theranostic efficiency of biosurfactants against COVID-19 and similar viruses - A review, 2022, 76, 17732247, 103764, 10.1016/j.jddst.2022.103764
    77. O. V. Kisil, V. S. Trefilov, V. S. Sadykova, M. E. Zvereva, E. A. Kubareva, Surfactin: Its Biological Activity and Possibility of Application in Agriculture, 2023, 59, 0003-6838, 1, 10.1134/S0003683823010027
    78. John Adewole Alara, Oluwaseun Ruth Alara, 2024, 9780443132889, 307, 10.1016/B978-0-443-13288-9.00001-2
    79. John Adewole Alara, 2024, 9780443132889, 355, 10.1016/B978-0-443-13288-9.00022-X
    80. Ines Ayadi, Sarra Akermi, Mariem Louati, Ali Gargouri, Lotfi Mellouli, Mohamed Guerfali, Microbial bioactive compounds from oleaginous yeast culture: insights into molecular docking interactions and toxicity prediction, 2024, 2190-6815, 10.1007/s13399-024-05516-2
    81. Swapnashree Satapathy, Ananya Kuanar, Arun Kumar Pradhan, 2024, 9789815196924, 80, 10.2174/9789815196924124010007
    82. Md. Sekendar Ali, Eon-Bee Lee, Walter H. Hsu, Kyoungho Suk, Syed Al Jawad Sayem, H. M. Arif Ullah, Seung-Jin Lee, Seung-Chun Park, Probiotics and Postbiotics as an Alternative to Antibiotics: An Emphasis on Pigs, 2023, 12, 2076-0817, 874, 10.3390/pathogens12070874
    83. Subhodip Mondal, Animesh Acharjee, Bidyut Saha, Utilization of natural surfactants: An approach towards sustainable universe, 2022, 60, 2572-8288, 1, 10.1002/vjch.202100137
    84. Mariana Amaral Azevedo, Letícia Portugal do Nascimento, Maria dos Remédios Vieira-Neta, Iolanda Cristina Silveira Duarte, 2023, Chapter 6, 978-3-031-31229-8, 129, 10.1007/978-3-031-31230-4_6
    85. Madduri Madhuri, Shivaprakash M. Rudramurthy, Utpal Roy, Two promising Bacillus-derived antifungal lipopeptide leads AF4 and AF5 and their combined effect with fluconazole on the in vitro Candida glabrata biofilms, 2024, 15, 1663-9812, 10.3389/fphar.2024.1334419
    86. O. V. Kisil, V. S. Trefilov, V. S. Sadykova, M. E. Zvereva, Е. А. Kubareva, SURFACTIN: BIOLOGICAL ACTIVITY AND THE POSSIBILITY OF AGRICULTURE APPLICATION (REVIEW), 2023, 59, 0555-1099, 3, 10.31857/S0555109923010026
    87. Fatimazahra Kadiri, Abdelkarim Ezaouine, Mohammed Blaghen, Faiza Bennis, Fatima Chegdani, Antibiofilm potential of biosurfactant produced by Bacillus aerius against pathogen bacteria, 2024, 56, 18788181, 102995, 10.1016/j.bcab.2023.102995
    88. Twinkle Rout, Muchalika Satapathy, Pratyasha Panda, Sibani Sahoo, Arun Kumar Pradhan, 2024, 9789815196924, 54, 10.2174/9789815196924124010005
    89. Avinash P. Ingle, Shreshtha Saxena, Mangesh Moharil, Mahendra Rai, Silvio S. Da Silva, 2023, 9781119854364, 173, 10.1002/9781119854395.ch9
    90. Karolína Englerová, Radomíra Nemcová, Eva Styková, Biosurfactants and their role in the inhibition of the biofilmforming pathogens, 2018, 67, 12107816, 107, 10.36290/csf.2018.015
    91. Francine Melise dos Santos, Amanda Pasinato Napp, Carolina Pinto de Aguiar, William Lautert Dutra, Breno GONÇALVES, Felipe Dalla Vecchia, Raj Deo Tewari, Clarissa Melo, Carbon Dioxide as a Carbon Source for Biosurfactant Production, 2025, 1556-5068, 10.2139/ssrn.5066224
    92. Priya Yadav, Rahul Prasad Singh, Subhasha Nigam, Rajpal Srivastav, Rachana Singh, Abeer Hasem, Elsayed Fathi Abd_Allah, Amit Raj, Sandeep Kumar Singh, Ajay Kumar, 2025, 24689289, 10.1016/bs.apmp.2024.11.002
    93. Roshan Jaiswal, Padmanaban Velayudhaperumal Chellam, Rangabhashiyam Selvasembian, Critical review on revamping circular economy strategies for the co-production of biosurfactants and lipase from agro-industrial wastes through resource recovery and life cycle assessment, 2025, 196, 09619534, 107733, 10.1016/j.biombioe.2025.107733
  • Reader Comments
  • © 2018 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(5698) PDF downloads(901) Cited by(5)

Figures and Tables

Tables(7)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog