Review

Prior art for the development of a fourth purification stage in wastewater treatment plant for the elimination of anthropogenic micropollutants-a short-review


  • The ubiquitous presence of pharmaceuticals in the aquatic environment is a worldwide problem of today. Current analytical methods reveal ever more pharmaceuticals in different water bodies. The concentrations in waters that are detectable reach nano-and picogram ranges, i.e., ppb and ppt levels. Observed concentrations amount to as high as few micrograms. Among the major entry paths are wastewater treatment plants, which are often unable to eliminate the pharmaceuticals sufficiently relying on their three conventional purification stages. Hence, pharmaceuticals enter the aquatic environment without desirable deconstruction. Thus, advanced wastewater treatment processes are under development to retain or eliminate these trace contaminants. According to Decision 2018/840, a watchlist of 15 contaminants of significant interest has been established for the monitoring of surface waters in the European Union. The contaminants include biocides and pharmaceuticals, among them three estrogens, estrone (E1), 17-β-estradiol (E2) and 17-α-ethinylestradiol, (EE2), the antibiotics azithromycin, clarithromycin and erythromycin of macrolide type, ciprofloxacin and amoxicillin of fluoroquinolone and betalactame type. This review will provide an overview of the currently explored and researched methods for the realization of a fourth purification stage in wastewater treatment plants. To this purpose, biological, chemical and physical purification processes are reviewed and their characteristics and potential discussed. The degradation efficacy of the pharmaceuticals on the EU-Watch list will be compared and evaluated with respect to the most promising processes, which might be realized on large scale. Last but not least, recent and novel pilot plants will be presented and discussed.

    Citation: Melanie Voigt, Alexander Wirtz, Kerstin Hoffmann-Jacobsen, Martin Jaeger. Prior art for the development of a fourth purification stage in wastewater treatment plant for the elimination of anthropogenic micropollutants-a short-review[J]. AIMS Environmental Science, 2020, 7(1): 69-98. doi: 10.3934/environsci.2020005

    Related Papers:

    [1] Tej Bahadur Shahi, Cheng-Yuan Xu, Arjun Neupane, William Guo . Machine learning methods for precision agriculture with UAV imagery: a review. Electronic Research Archive, 2022, 30(12): 4277-4317. doi: 10.3934/era.2022218
    [2] Abul Bashar . Employing combined spatial and frequency domain image features for machine learning-based malware detection. Electronic Research Archive, 2024, 32(7): 4255-4290. doi: 10.3934/era.2024192
    [3] Xiaomeng An, Sen Xu . A selective evolutionary heterogeneous ensemble algorithm for classifying imbalanced data. Electronic Research Archive, 2023, 31(5): 2733-2757. doi: 10.3934/era.2023138
    [4] Yixin Sun, Lei Wu, Peng Chen, Feng Zhang, Lifeng Xu . Using deep learning in pathology image analysis: A novel active learning strategy based on latent representation. Electronic Research Archive, 2023, 31(9): 5340-5361. doi: 10.3934/era.2023271
    [5] Mashael S. Maashi, Yasser Ali Reyad Ali, Abdelwahed Motwakel, Amira Sayed A. Aziz, Manar Ahmed Hamza, Amgad Atta Abdelmageed . Anas platyrhynchos optimizer with deep transfer learning-based gastric cancer classification on endoscopic images. Electronic Research Archive, 2023, 31(6): 3200-3217. doi: 10.3934/era.2023162
    [6] Jiawang Li, Chongren Wang . A deep learning approach of financial distress recognition combining text. Electronic Research Archive, 2023, 31(8): 4683-4707. doi: 10.3934/era.2023240
    [7] Ju Wang, Leifeng Zhang, Sanqiang Yang, Shaoning Lian, Peng Wang, Lei Yu, Zhenyu Yang . Optimized LSTM based on improved whale algorithm for surface subsidence deformation prediction. Electronic Research Archive, 2023, 31(6): 3435-3452. doi: 10.3934/era.2023174
    [8] Shixiong Zhang, Jiao Li, Lu Yang . Survey on low-level controllable image synthesis with deep learning. Electronic Research Archive, 2023, 31(12): 7385-7426. doi: 10.3934/era.2023374
    [9] Xiao Chen, Fuxiang Li, Hairong Lian, Peiguang Wang . A deep learning framework for predicting the spread of diffusion diseases. Electronic Research Archive, 2025, 33(4): 2475-2502. doi: 10.3934/era.2025110
    [10] Jianting Gong, Yingwei Zhao, Xiantao Heng, Yongbing Chen, Pingping Sun, Fei He, Zhiqiang Ma, Zilin Ren . Deciphering and identifying pan-cancer RAS pathway activation based on graph autoencoder and ClassifierChain. Electronic Research Archive, 2023, 31(8): 4951-4967. doi: 10.3934/era.2023253

  • The ubiquitous presence of pharmaceuticals in the aquatic environment is a worldwide problem of today. Current analytical methods reveal ever more pharmaceuticals in different water bodies. The concentrations in waters that are detectable reach nano-and picogram ranges, i.e., ppb and ppt levels. Observed concentrations amount to as high as few micrograms. Among the major entry paths are wastewater treatment plants, which are often unable to eliminate the pharmaceuticals sufficiently relying on their three conventional purification stages. Hence, pharmaceuticals enter the aquatic environment without desirable deconstruction. Thus, advanced wastewater treatment processes are under development to retain or eliminate these trace contaminants. According to Decision 2018/840, a watchlist of 15 contaminants of significant interest has been established for the monitoring of surface waters in the European Union. The contaminants include biocides and pharmaceuticals, among them three estrogens, estrone (E1), 17-β-estradiol (E2) and 17-α-ethinylestradiol, (EE2), the antibiotics azithromycin, clarithromycin and erythromycin of macrolide type, ciprofloxacin and amoxicillin of fluoroquinolone and betalactame type. This review will provide an overview of the currently explored and researched methods for the realization of a fourth purification stage in wastewater treatment plants. To this purpose, biological, chemical and physical purification processes are reviewed and their characteristics and potential discussed. The degradation efficacy of the pharmaceuticals on the EU-Watch list will be compared and evaluated with respect to the most promising processes, which might be realized on large scale. Last but not least, recent and novel pilot plants will be presented and discussed.



    Common macular and vascular diseases include age-related macular degeneration (ARMD), diabetic macular edema (DME), branch retinal vein occlusion (BRVO), central retinal vein occlusion (CRVO), and central serous chorioretinopathy (CSCR), which are the leading causes of visual impairment and blindness worldwide [1,2,3]. According to the World Health Organization (WHO), DME, which primarily affects working-age adults, affected 425 million people worldwide in 2017 and is expected to affect 629 million people by 2045 [4]. The WHO also estimates that 196 million people had ARMD in 2020; this number is expected to rise to 288 million by 2040 [5]. The prevalence of ARMD in elderly people is 40% at the age of 70 years, rising to 70% at the age of 80 years. Rogers et al. [6] discovered that BRVO and CRVO affected 13.9 million and 2.5 million of the world's population aged 30 years and older, respectively, in 2008. Men have a higher prevalence of CSCR compared to women [7]. A large population is afflicted by these diseases, and projections suggest that this number will escalate in the future. However, the first stage of these diseases can be treated, and patients can recover their vision loss through early detection and treatment [8,9,10].

    Optical coherence tomography (OCT) is a noninvasive imaging modality that provides high-resolution information within a cross sectional area. OCT retinal imaging enables visualization of the thickness, structure, and detail of various layers of the retina. In addition, when the retina develops a disease, OCT enables the visualization of abnormal features and damaged retinal structures [11]. Therefore, retinal OCT images are widely used in the medical field to monitor information in medical images prior to treatment or for the diagnosis of various diseases.

    For several years, ophthalmologists have analyzed the comprehensive information inside the retina for retinal care services, treatment, and diagnosis using retinal OCT images in clinical settings. The clinician performs these tasks manually and wait for each process. As a result, manual analysis is time consuming when there are numerous OCT images. Even if the clinician has great expertise, this analysis may not be accurate [12]. An automated technique based on deep learning (DL) or machine learning using artificial intelligence has been proposed as a solution to overcome this limitation.

    Recently, computer algorithms based on artificial intelligence, DL, and machine learning have been proposed for the automatic diagnosis of various retinal diseases and have been applied in clinical health care. Han et al. [13] modified three well known convolutional neural network (CNN) models to gain access to normal and three subtypes of neovascular age-related macular degeneration (nAMD). The classification layers of the original CNN models were replaced by new layers: four fully connected layers and three dropout layers, along with a Leaky rectified linear activation unit (ReLU) as an activation function. The modified models were trained using the transfer learning technique and tested on 920 OCT images; the VGG-16 model achieved an accuracy of 87.4%. Sotoudeh-Paima et al. [14] classified OCT images to identify normal, AMD, and choroidal neovascularization (CNV) using a multiscale CNN. This CNN was evaluated and achieved a classification accuracy of 93.40% on the public dataset. Elaziz et al. [15] developed a four-class classification method for accessing retinal diseases from OCT images based on an ensemble DL model and machine learning. First, the features are extracted from the two models, MobileNet and DenseNet, and were concatenated as full features of the input images. Then, feature selection was performed to remove irrelevant features and to input the useful features into machine learning to classify the input data. A total of 968 OCT images were used to evaluate classification performance, and an accuracy of 94.31% was achieved. Another study by Liu et al. [16] used a DL model to extract attention features from OCT images. It used the extracted features as guiding features for CNV, DME, drusen, and normal. The classification performance was assessed using public datasets, and an average accuracy of 95.10% was achieved. Minagi et al. [17] used transfer learning with universal adversarial perturbations (UAPs) for classification with a limited dataset. Three types of medical images, including OCT images, were used to assess diseases, and DL models were trained using the ImageNet dataset. The UAPs algorithm was used to generate a training set based on the data provided to train the DL model. There were 11,200 OCT images utilized in training and assessing the model's performance, and a classification accuracy of 95.3% was achieved for the four classes: CNV, DME, drusen, and normal. Tayal et al. [18] presented four ocular disease classifications based on three CNN models using OCT images. Images were enhanced before being fed to CNN models. To assess the performance of the presented method, 6,678 publicly available OCT images were evaluated. The method achieved an accuracy of 96.50% with a CNN model which compressed nine layers. The performance of the CNN models with nine layers outperformed the experimented CNN models with five and seven layers. Adversarial retraining is an algorithm used to improve the performance of DL models based on classification.

    According to the literature, retinal OCT classification was developed using DL and DL based methods such as transfer learning, smoothing generative adversarial networks, adversarial retraining, and multi-scale CNN. This method is used to improve the model's performance by fine-tuning previous task knowledge using the OCT image problem, increasing the dataset size for training, applying the technique of inputting data for the training model, and changing the training input image sizes. However, the classification methods can achieve an accuracy of less than 97.00%, indicating their potential for further improvement. Moreover, these studies classify retinal diseases into fewer than five classes. This study aims to improve the classification accuracy and detect five classes of retinal diseases, which are more than the previous studies highlighted in the literature.

    In this study, we propose an automatic method based on a hybrid of deep learning and ensemble machine learning for screening five different retinal diseases from OCT images to improve the performance of OCT image classification. The proposed method improves classification accuracy, outperforming standalone classifiers without a hybrid. In addition, it can be trained using a smaller dataset from our hospital, which has been strictly labelled by experts. Moreover, the proposed method enables deployment with a web server for open access to test the evaluation performance within seconds.

    All OCT images were collected from Soonchunhyang University's Bucheon Hospital. The OCT images were collected and normalized after approval by the Bucheon Hospital's Institutional Review Board (IRB). OCT images were captured using DRI-OCT (Topcon Medical System, Inc., Oakland, NJ, USA). The scan range was 3–12 mm in the horizontal and vertical directions, with a lateral resolution of 20 μm and an in-depth resolution of 8 μm. The shooting speed was 100,000 A-scans per second. The OCT images utilized were collected twice; the first comprised 2,000 images that were captured between April and September 2021, while the second consisted of 998 images, and took place over a period of approximately five months from September 2021 to January 2022. Therefore, the total number of OCT images collected twice was 2,998; these were labeled by ophthalmologists for five retinal diseases (ARMD:740, BRVO:450, CRVO:299, CSCR:749, and DME:760) as the ground truth.

    This study was approved by the Institutional Review Board (IRB) from Soonchunhyang University Bucheon Hospital, Bucheon, Republic of Korea (IRB approval number: 2021-05-001). All methods were performed in accordance with relevant guidelines and regulations. Informed consent was obtained from all subjects.

    Image processing is a technique for performing various operations on the original images to convert it into a format suitable for DL models or to extract useful features. In image classification based on deep learning, image processing is an essential initial process to change an image before feeding it to the CNN model. The CNN model requires a unique size for the image input, and higher-resolution images demand longer computing times. To shorten the operating time and the suitable size required by the CNN models, all OCT images were downsized to 300 pixels in height and 500 pixels in width. The OCT image dataset was split into an 80% training set and 20% testing set. The training set was used to train the deep learning model and the testing set was used to assess performance.

    The size of the dataset has a significant impact on the DL performance. Therefore, a larger dataset may enable a better performance. However, in the medical field, most medical dataset has size limits. Data augmentation is a technique developed to overcome the limitations of a dataset by performing different operations on the data provided and creating new data, thereby enhancing the dataset size. Additionally, data augmentation is used to enhance performance [19], generalize the model [20], and avoid overfitting [21]. We utilize data augmentation techniques from the Python library imgaug including like vertical flip, rotation, scale, brightness, saturation, contrast, enhance and contrast, and equalization. The OCT images were augmented at angles of 170,175,185, and 190. The selected angle is suitable for rectangle shape representation without loss of information from the original OCT images; scale image with a random range between 0.01 to 0.12; the level of brightness from 1 to 3; the saturate operation, which ranges from 1 to 5, increases by one with each level; random contrast with contrast values ranging from 0.2 to 3; enhance and contrast with levels ranging from 1 to 1.5; and image equalization with levels ranging from 0.9 to 1.4. At the end of the data augmentation process, one OCT image can serve as the basic for generating 29 augmented images. Therefore, the training set comprised a total of 69,455 OCT images, including samples. The acquired OCT and augmented images are shown in Figure 1. Applying data augmentation, only the training set is used for training the proposed method. After finishing the augmentation operation, the OCT images are passed through the 10-fold cross-validation technique to partition the data into folds for the training model (training data) and to test the model after finishing every epoch (validation data).

    Figure 1.  OCT images before and after performing data augmentation. (a) represents the original OCT image. (b), (c), and (d) illustrate brightness adjustments. (e) and (f) demonstrate contrast modifications. (g), (h), and (i) display contrast enhancement. (j) and (k) depict equalization. (l) represents a vertical flip. (m), (n), (o), (p), (q), and (r) indicate angle rotations. (s), (t), (u), (v), and (w) illustrate saturation changes. (x), (y), (z), (A), (B), and (C) represent scaling variations.

    Figure 2 shows the architecture of the proposed method that comprises three significant blocks: feature extraction, classification, and boosting performance. First, transfer learning based on CNN models extracts one thousand features from the OCT images. Second, various machine learning algorithms are used to classify the OCT images based on the features extracted by the CNN model. Finally, the ensemble algorithm fuses the distribution probabilities of the same class and predicts the retinal disease class based on probability. Each block of the proposed architecture is described in detail in the following subsections.

    Figure 2.  System architecture overview of the proposed method. The proposed method accepts images with resolution of 500 pixels in width and 300 pixels in height. CNN models extract features from OCT images and classify them using machine learning algorithms. Voting classifier ensemble output probabilities for predicting retinal disease.

    Transfer learning is a technique used to transform the knowledge of a related task that has already been studied to improve the learning of a new task. Training a CNN model from scratch is computationally expensive and time consuming; moreover, an extensive dataset is required to achieve a better performance. Therefore, transfer learning has been developed to overcome DL's drawbacks [22]. To retrain the model with new tasks based on prior knowledge, pretrain was refined, small top layers were trained, and the final layers were frozen. In this study, the transfer learning CNN (TL-CNN) models EfficientNetB0 [23], InceptionResNetV2 [24], InceptionV3 [25], ResNet50 [26], VGG16 [27], and VGG19 [28] are selected and updated. The modification names of the CNN models start with TL, indicating transfer learning, and ends with the original names of the CNN models, including TL-EfficientNetB0, TL -InceptionResNetV2, TL-InceptionV3, TL-ResNet50, TL-VGG16, and TL-VGG19. The original CNN models were created for generic image classification tasks. They were trained and tested on a large dataset (ImageNet) to categorize 1000 different types of images. To use a CNN model with the transfer learning technique and classify retinal OCT images, each CNN model must modify its classification layers to adapt to the target classes. One specific problem is the categorization of OCT images. The new classification layer is modified with continued stacking of GlobalAveragePooling2D, one Normalization layer, and two Dense layers. The first Dense layer consists of 1,024 with the ReLU activation function and the final dense layer with a five output- vectors. Finally, the updated model is pretrained, and the pretrain model is retrained to fine-tune the previous feature representation in the base model to make it more relevant for OCT image classification. The output consists of five vectors representing the distribution class probabilities using the Softmax activation function. As mentioned previously, a CNN model based on transfer learning is used to extract convolutional features from the OCT images. Therefore, the convolutional features were extracted from the TL-CNNs models where the GlobalAveragePooling2D layers of the classification layer. These features are one-dimensional. Different models provide various features and numbers based on the structure and convolution filters of the model.

    Six TL-CNN models independently extracted the features. At the GlobalAverage-Pooling2D layers, the features were extracted (TL-EfficientNetB0: 1,280 features, TL-InceptionResnetV2: 1,536 features, TL-InceptionV3: 2,048 features, TL-ResNet50: 2,048 features, TL-VGG16: 512 features, TL-VGG19: 512 features). Then, the extracted features of each TL-CNN model were used as the input to five popular machine learning classifiers: support vector machine (SVM) [29], k-nearest neighbors (k-NN) [30], decision tree (DT) [31], Random Forests (RF) [32], Naïve Bayes [33], and XGBoost [34]. Various machine learning classifiers use different techniques for learning and distinguishing the different classes of data.

    Individual machine learning classifiers provide different identification accuracies. This is because each classifier has its own learning ability to identify classes based on the given features. Therefore, an ensemble method is used to aggregate the distribution probabilities of the two classifiers. The proposed method selects two higher prediction classifiers (k-NN and XGBoost) based on an experiment to perform aggregation. An ensemble is a type of soft voting that performs better than other models [35]. Soft voting predicts the final class label as the class label most frequently predicted by classifiers. In soft voting, class labels are predicted by averaging the probability p of the class. Table 1 presents the proposed algorithm, which includes image processing, splitting data, data augmentation, feature extraction, classification, and an ensemble of classifiers:

    yFC=argmaximk=1wkpik (1)
    Table 1.  Algorithm of the proposed method.
    Algorithm 1: Proposed OCT images Classification
    1: procedure OCT IMAGES PROCESSING
    2:       return preprocessed-images

    3: procedure SPLIT-DATA (OCT-data)
    4: train-data, test-data, train-labels, test-labels = split (OCT-images, labels)

    5: procedure DATA AUGMENTATION (train-data)
    6: augmented images = augmentation (vertical flip, rotation, scale, brightness, saturation, contrast, enhance and
    7: contrast, and equalization)
    8:       return augmented-images

    9: procedure 10-FOLD_CROSS_VALIDATION (augmented images, labels)
    10: Fold1, Fold2, ……Fold10 = train_test_split(augmented images, labels)
    11:       return Fold1-10

    12: procedure FEATURE_EXTRACTION (Fold1-10, test-data, test-labels)
    13: TL-CNN models = modify the convolutional neural network (CNN) models
    14: pre-train the TL-CNN models, small top layers are trained, and the final layers are frozen.
    15: extracted features = TL-CNN model at GlobalAveragePooling2D layers
    16: return extracted features saved in csv format
    17: procedure CLASSIFICATION (extracted features, labels)
    18: classifiers = [svm, k-NN, DT, RF, Naïve-Bayes, and XGBoost]
    19:   for clsf in range (0, 6):
    20:       predicted-labels = classifiers[clsf]. fit (extracted-features)
    21:       training-accuracy = accuracy (predicted-labels, labels)
    22:       save_train_weight
    23: voting = "soft"
    24: ML1 = k-NN (train-data, train-labels, test-data)
    25: ML2 = XGBoost (train-data, train-labels, test-data)
    26: procedure ENSMEBLE_CLASSIFIERS (train-data, train-labels, test-data)
    27: ensemble-classifiers = concadenate(ML1, ML2)
    28: ensemble-classifers.fit (train-data, train-labels)
    29: predictions = ensemble-classifers.predict(test-data)
    30: save_training_weights, results_visualization

     | Show Table
    DownLoad: CSV

    where w_k is the weight of the machine learning classifiers, which can be either k-NN or XGBoost; it automatically learns from disease features in OCT images and then identifies the type of disease based on the input data; i represents the class label of the retinal diseases, where i ∈{0: ARMD, 1: BRVO, 2: CRVO, 3: CSCR, 4: DME}; and p_ik represents the probability of machine-learning weight k for class i.

    The proposed OCT image classification method was developed using Python 3.7 and TensorFlow 2.6.0. In addition, Scikit Learn was operated on a personal computer running the Windows 10 operating system powered by an Intel(R) Xeon (R) Silver 4114 @ 2.20GHz CPU, 192GB RAM, and an NVIDIA TITAN RTX 119GB GPU.

    The proposed OCT image classification method was trained using augmented OCT images and evaluated using a test set. There were two types of training. First, six TL-CNN models were trained to perform feature extraction from OCT images.

    Six TL-CNN models were separately trained using a combination of the training set and the augmented images of the training set. The combination data were split using a 10-fold cross-validation algorithm to separate the images for training, validate the model during training, and prevent overfitting. Furthermore, the TL-CNN models were individually trained with a fixed batch size of 64, epochs of size 100, and an Adam optimizer with a learning rate of 0.0001. The learning rate was selected based on the standard learning rate provided by the TensorFlow library. For example, with a setting of 100 epochs, each model must be trained 100 times on the same data. Therefore, the performance is improved by updating the weight based on the information lost through repetitions of a training session. The weights of each TL-CNN model were recorded in a separate file after training and were utilized to extract features from the training and testing sets. Then, the machine learning models were trained with the convolution features extracted by the TL-CNN models to access the class probabilities. Six machine learning models were separately trained, and the weights were recorded after the training completed. Finally, an ensemble method based on soft voting was applied to the average class probabilities of the two classifiers to obtain an effective final class prediction.

    The results of the proposed OCT image classification are divided into three parts: classification results, deployment of the classification results to web services, and a comparison of the results with similar studies in terms of classification accuracy.

    A test set was used to evaluate the performance of the proposed method after training the model. The same preprocessing was performed on both the test dataset and the training dataset without data augmentation. The test set contained 601 OCT images, which were used to assess the classification performance. Six TL-CNN models were individually trained to extract features from the OCT images and store the extracted features in pickle format. Six machine learning classifiers were utilized to discriminate the classes of OCT images based on the features extracted by the TL-CNN. Statistical theories were analyzed to measure the classification ability among the classes, sensitivity, specificity, precision, and accuracy. The relationship between the sensitivity and specificity of various categories was shown through a receiver operating characteristic (ROC) curve. Moreover, the confusion matrix was analyzed, which indicated the correct and incorrect class predictions. Table 2 lists the test results of using TL-EfficientNetB0 as an extractor and seven types of classifiers, including an ensemble classifier, the classification result outperformed the ensemble classifier with a sensitivity, specificity, precision, and accuracy of 96.17, 98.92, 95.89 and 95.85%, respectively. The second highest performance was achieved with the k-NN classifier, which achieved a sensitivity, specificity, precision, and accuracy of 87.37, 96.95, 88.82 and 88.89%, respectively. The classification results for the other machine learning classifiers are unstable, both increasing and decreasing randomly.

    Table 2.  Shown are OCT images before and after performing data augmentation. (a) represents the original OCT image. (b), (c), and (d) illustrate brightness adjustments. (e) and (f) demonstrate contrast modifications. (g), (h), and (i) display contrast enhancement. (j) and (k) depict equalization. (l) represents a vertical flip. (m), (n), (o), (p), (q), and (r) indicate angle rotations. (s), (t), (u), (v), and (w) illustrate saturation changes. (x), (y), (z), (A), (B), and (C) represent scaling variations.
    TL-CNN model Machine Learning Sensitivity Specificity Precision Accuracy
    TL-EfficientNetB0 SVM 86.79% 96.78% 88.64% 88.39%
    k-NN 87.37% 96.95% 88.82% 88.89%
    DT 85.80% 96.41% 84.95% 86.90%
    RF 66.20% 92.60% 81.08% 75.95%
    Naive Bayes 86.11% 96.40% 86.58% 86.90%
    XGBoost 85.86% 96.45% 86.38% 87.23%
    Ensemble 96.17% 98.92% 95.89% 95.85%

     | Show Table
    DownLoad: CSV

    Table 3 shows the classification results when using TL-InceptionResnetV2 as an extractor and seven classifiers, showing that the result outperforms the ensemble classifier with a sensitivity, specificity, precision, and accuracy of 97.42, 99.40, 97.49 and 97.68%, respectively. The second highest performance was achieved with the k-NN classifier, with a sensitivity, specificity, precision, and accuracy of 87.37, 96.48, 88.19 and 87.56%, respectively. In addition, with the same extractor, the classification performance of XGBoost was similar to that of the k-NN classifier. Table 4 lists the evaluation results when using the TL-InceptionV3 extractor and seven machine learning classifiers, including the ensemble classifier, which outperformed other methods with a sensitivity, specificity, precision, and accuracy of 91.34, 97.59, 91.03 and 91.04%, respectively. The second highest performance was achieved by XGBoost, with a sensitivity, specificity, precision, and accuracy of 84.42, 95.10, 82.88, and 82.91%, respectively. Table 5 lists the classification results when using the TL-ResNet50 model as a feature extractor and classifying those features by seven different classifiers, which indicates that using ensemble classifiers outperforms the obtained a sensitivity, specificity, precision, and accuracy of 96.46, 99.14, 96.76 and 96.68%, respectively. The second highest performance was achieved by XGBoost, with a sensitivity, specificity, precision, and accuracy of 87.63, 96.59%, 88.27 and 87.73%, respectively. The performances of the other two classifiers, SVM and k-NN, were comparable and better than those of the three classifiers in the experiments. Table 6 lists the test results of the proposed classification with VGG-16 as a feature extractor and seven machine learning classifiers, the ensemble classifier exhibited the best performance, with a sensitivity, specificity, precision, and classification accuracy of 92.07, 98.00, 92.60 and 92.54%, respectively. The XGBoost classifier had the second highest performance for TL-VGG16 as a feature extractor; it obtained a sensitivity, specificity, precision, and accuracy of 80.48, 94.91, 81.44 and 82.26%, respectively. A similar performance was observed for SVM and k-NN. Table 7 lists the classification test results of the TL-VGG19 model for feature extraction and classification using these features by various classifiers. Ensemble classifiers algorithm outperformed the five other classifiers; its sensitivity, specificity, precision, and accuracy are 93.86, 93.40, 93.44 and 93.86%, respectively. The second- and third-highest performances were achieved by XGBoost and SVM, respectively.

    Table 3.  Performance summary of proposed classification through feature extraction using TL-InceptionResnetV2, six classifiers, and ensemble voting classifiers. Various sensitivities, specificities, precisions, and accuracies are obtained using different classifiers. The proposed classification method with ensemble classifiers outperforms all statistic measurements.
    TL-CNN model Machine Learning Sensitivity Specificity Precision Accuracy
    TL-InceptionResnetV2 SVM 86.27% 96.13% 86.86% 86.40%
    k-NN 87.37% 96.48% 88.19% 87.56%
    DT 83.77% 95.54% 83.37% 84.41%
    RF 72.93% 93.79% 80.67% 79.27%
    Naive Bayes 79.66% 93.41% 78.25% 77.78%
    XGBoost 87.29% 96.47% 88.05% 87.56%
    Ensemble 97.42% 99.40% 97.49% 97.68%

     | Show Table
    DownLoad: CSV
    Table 4.  Performance summary of proposed classification through feature extraction using TL-InceptionV3, six classifiers, and ensemble voting classifiers. Various sensitivities, specificities, precisions, and accuracies are obtained when using different classifiers. The proposed classification method with ensemble classifiers outperforms all statistic measurements.
    TL-CNN models Machine Learning Sensitivity Specificity Precision Accuracy
    TL-InceptionV3 SVM 82.42% 94.50% 80.85% 81.09%
    k-NN 83.05% 94.61% 80.94% 81.43%
    DT 81.54% 94.18% 79.65% 80.09%
    RF 79.50% 94.01% 80.52% 79.77%
    Naive Bayes 65.72% 86.52% 2.58% 61.33%
    XGBoost 84.42% 95.10% 82.88% 82.91%
    Ensemble 91.34% 97.59% 91.03% 91.04%

     | Show Table
    DownLoad: CSV
    Table 5.  Performance summary of proposed classification through features extraction using TL-ResNet50, six classifiers, and ensemble voting classifiers. Various sensitivities, specificities, precisions, and accuracies are obtained using different classifiers. The proposed classification method with ensemble classifiers outperforms all statistic measurements.
    TL-CNN model Machine Learning Sensitivity Specificity Precision Accuracy
    TL-ResNet50 SVM 86.25% 96.02% 85.12% 85.95%
    k-NN 85.75% 96.00% 86.26% 85.74%
    DT 82.04% 94.94% 82.34% 82.59%
    RF 52.90% 88063% 71.77% 65.67%
    Naive Bayes 67.71% 89.82% 72.49% 64.68%
    XGBoost 87.63% 96.59% 88.27% 87.73%
    Ensemble 96.46% 99.14% 96.76% 96.68%

     | Show Table
    DownLoad: CSV
    Table 6.  Performance summary of proposed classification through features extraction using TL-VGG16, six classifiers, and ensemble voting classifiers. Various sensitivities, specificities, precisions, and accuracies are obtained using different classifiers. The proposed classification method with ensemble classifiers outperforms all statistic measurements.
    TL-CNN model Machine Learning Sensitivity Specificity Precision Accuracy
    TL-VGG16 SVM 76.39% 93.49% 76.96% 78.28%
    k-NN 74.55% 92.33% 75.70% 74.96%
    DT 57.42% 84.86% 55.72% 58.37%
    RF 50.55% 86.58% 38.39% 63.18%
    Naive Bayes 59.53% 85.08% 58.84% 59.20%
    XGBoost 80.48% 94.91% 81.44% 82.26%
    Ensemble 92.07% 98.00% 92.60% 92.54%

     | Show Table
    DownLoad: CSV
    Table 7.  Performance summary of proposed classification through features extraction using TL-VGG19, six classifiers, and ensemble voting classifiers. Various sensitivities, specificities, precisions, and accuracies are obtained using different classifiers. The proposed classification method with ensemble classifiers outperforms all statistic measurements.
    TL-CNN model Machine Learning Sensitivity Specificity Precision Accuracy
    TL-VGG19 SVM 79.90% 93.74% 78.77% 78.82%
    k-NN 69.16% 90.99% 70.56% 71.64%
    DT 53.23% 82.94% 53.41% 54.73%
    RF 48.29% 85.62% 37.71% 60.36%
    Naive Bayes 56.41% 82.96% 54.70% 54.89%
    XGBoost 82.44% 95.30% 81.90% 83.58%
    Ensemble 93.86% 93.40% 93.44% 93.86%

     | Show Table
    DownLoad: CSV

    Six TL-CNN models were compared, and TL-InceptionResNetV2 achieved a better performance than the other five models used in this study, with a sensitivity, specificity, precision, and accuracy of 97.42, 99.40, 97.49 and 97.68 %, respectively. The ensemble algorithm always outperformed all the TL-CNN models. The individual k-NN and XGBoost classifiers performed better than the three individual classifiers. Thus, ensembled k-NN and XGBoost also achieved better performance than k-NN and XGBoost.

    Figure 3 shows the ROC result of the proposed classification method, which outperforms TL- InceptionResnetV2 with ensemble classifiers (k-NN and XGBoost). The ROC among each class ARMD, BRVO, CRVO, CSCR, and DME is 0.99, 0.96, 0.99, 0.99, and 0.98, respectively. The relationship between sensitivity and specificity of the five classes is most important. The confusion matrix is implemented by using the Sklearn library in Python. The size of test data is essential to present the robustness of classification. The confusion matrix shows the number of correct and incorrect predictions among all classes. Figure 4 shows the confusion matrix of the proposed method which exhibited best performance; 148 of 149 OCT images of ARMD class are correctly predicted, 85 of 91 images of BRVO class are correctly predicted (ARMD:3 and DME:3 are incorrect predictions), 59 of 60 images are correctly predicted as CRVO and one image that is incorrectly predicted as BRVO, 148 of 150 images are correctly predicted and two images are incorrectly predicted as ARMD, and 149 of 153 are correctly predicted, and four are incorrectly predicted (ARMD: 1, BRVO:1, and CRVO:2 are incorrect prediction).

    Figure 3.  ROC curve of the proposed classification method, which exhibits best accuracy on TL-InceptionResnetV2 model and ensemble classifiers.
    Figure 4.  Confusion matrix of the proposed method when it exhibits best performance on TL-InceptionResNetV2 and ensemble classifiers.

    To render the proposed method applicable and accessible from outside through an Internet connection, we deployed the proposed OCT image classification to a web server using the Flask framework. The web server receives one image input at a time and inputs it into the proposed classification method to predict retinal diseases. The input image is an OCT image consisting of three channels with a resolution of 300 pixels in height and 500 pixels in width. When inputting an OCT image through a web service user interface (UI), the image is transferred to a computer server that runs a DL classification model. First, the computer server performs image processing which is the same to the processes used in both the train and test sets. Second, the preprocessed image is inputted into the proposed classification weights for prediction. Finally, the predicted results are forwarded to the web service using the Flask framework. The prediction results consist of the image input, distribution probabilities among the five classes, the retinal disease diagnosis class, and prediction times of an image. The prediction time is the time taken to input an image to a web service to predict and return the prediction result. Figure 5 shows the initial UI of the web server. The prediction results obtained after inputting the OCT images are shown in Figure 6.

    Figure 5.  Initial user interface of the developed web service for OCT image classification. The "Select an Image" button allows the user to browse to the location of a stored image and upload it to the webservice, and the "Predict" button sends the image to a deep learning server and receives the diagnosis class.
    Figure 6.  Prediction results from the development web service for OCT image classification. The predicted OCT image, distribution probabilities among five classes of retinal diseases in percent, a final predicted class based on higher probability, and time prediction are represented.

    The higher accuracy of the proposed OCT image classification method is compared with that of the recent studies reviewed in the literature review section, as listed in Table 8. These studies focused on transfer learning, developing new models, and combining well known CNN models with machine learning. All the listed studies used either different OCT databases or a combination of these datasets. Moreover, the number and type of classification classes were different, with at most four classes. We classify retinal diseases into five classes using a dataset obtained from a hospital. An additional number of classes can affect the performance of the classification methods. Table 8 lists the methods and algorithms that have been presented, including the suggested model with transfer learning, the multiscale DL model, and transfer learning using existing CNN models. However, the results as listed in the literature review have shown an accuracy of < 97%. Instead of focusing on a single classifier, this study combines two machine-learning classifiers and the DL as a feature extractor. Our study exhibits an accuracy of 97.68%, which is greater than the accuracy of the aforementioned studies. In addition, the number of classification classes is higher than that of the studies reviewed.

    Table 8.  Results comparison.
    Author Year Method Disease type Dataset size Accuracy
    Han et al. [13] 2022 Transfer learning with a modification of the well-known CNN models 4-class: PCV, RAP, nAMD, and NORMAL 4749 87.4%
    Sotoudeh-Paima et al. [14] 2022 Deep learning: multi-scale convolutional neural network 3-class: AMD, CNV, NORMAL 120,961 93.4%
    Elaziz et al. [15] 2022 Ensemble deep learning model for feature extraction, features selection, machine learning as classifier. 4-class: DME, CNV, DRUSEN, and NORMAL 84,484 94.32%
    Liu et al. [16] 2022 Deep learning based on method and lesions segmentation model. 4-class: CNV, DME, DRUSEN, and NORMAL 86,134 95.10%
    Minagi et al. [17] 2022 Transfer learning with DNN models 4-class: CNV, DME, DRUSEN, and NORMAL 11,200 95.3%
    Tayal et al. [18] 2022 Deep learning-based method 4-class: DME, CNV, DRUSEN, NORMAL 84,484 96.5%
    Proposed method Hybrid of deep learning and machine learning + ensemble machine learning classifiers. 5-class: ARMD, BRVO, CRVO, CSCR, DME 2,998 97.68%

     | Show Table
    DownLoad: CSV

    Our study classifies retinal OCT images with disease classes that differ from the reviewed studies and are not available in the public dataset. We hope that these retinal diseases will become available in the future, and we will evaluate the proposed OCT image classification system using a public dataset.

    This study presents a hybrid ensemble OCT image classification method for the diagnosis of five classes of retinal diseases. The proposed method employs an ensemble machine learning classifier as the classifier and a hybrid deep learning model as the feature extractor. We identified the deep learning model and ensemble classifiers that were most suitable for OCT image classification. The proposed model outperformed an individual classifier. With an accuracy of 97.68%, the best deep learning model and ensemble machine learning classifiers of the proposed method were TL- InceptionResnetV2 and the aggregation of KNN and XGBoost. This classification can be deployed to web services for convenient access to diagnose retinal disease from outside the Internet. Moreover, the prediction time in seconds was short, reducing the time required for prediction. This study contributes to the development of accurate multiclass OCT image classification. In the future, we aim to improve the classification performance. If datasets with the same class as in our study are made public, we will assess the proposed method on these datasets to broaden their applicability. In the medical field, improved performance can be used to automatically classify OCT images and eliminate time-consuming tasks, and this classification can also aid in the prevention of vision loss.

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

    The data used to support this study have not been made available access because they are real clinical data from Soonchunhyang Bucheon Hospital, and patient's privacy should be protected, it enables to detect people through this data, but they are available from the corresponding author on reasonable request.

    This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. NRF-2021R1A2C1010362) and the Soonchunhyang University Research Fund.

    The authors declare no competing interests.



    [1] Tijani JO, Fatoba OO, Petrik LF (2013) A review of pharmaceuticals and endocrine-disrupting compounds: Sources, effects, removal, and detections. Water Air Soil Pollut 224.
    [2] Gerbersdorf SU, Cimatoribus C, Class H, et al. (2015) Anthropogenic Trace Compounds (ATCs) in aquatic habitats-Research needs on sources, fate, detection and toxicity to ensure timely elimination strategies and risk management. Environ Int 79: 85-105. doi: 10.1016/j.envint.2015.03.011
    [3] Burke V, Richter D, Greskowiak J, et al. (2016) Occurrence of Antibiotics in Surface and Groundwater of a Drinking Water Catchment Area in Germany. Water Environ Res 88: 652-659. doi: 10.2175/106143016X14609975746604
    [4] Schwarzenbach RP (2006) The Challenge of Micropollutants in Aquatic Systems. Science 313: 1072-1077. doi: 10.1126/science.1127291
    [5] Luo Y, Guo W, Ngo HH, et al. (2014) A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment. Sci Total Environ 473-474: 619-641. doi: 10.1016/j.scitotenv.2013.12.065
    [6] Haddad T, Baginska E, Kümmerer K (2015) Transformation products of antibiotic and cytostatic drugs in the aquatic cycle that result from effluent treatment and abiotic/biotic reactions in the environment: An increasing challenge calling for higher emphasis on measures at the beginning of the pi. Water Res 72: 75-126. doi: 10.1016/j.watres.2014.12.042
    [7] Rivera-Utrilla J, Sánchez-Polo M, Ferro-García MÁ, et al. (2013) Pharmaceuticals as emerging contaminants and their removal from water. A review. Chemosphere 93: 1268-1287. doi: 10.1016/j.chemosphere.2013.07.059
    [8] Kümmerer K (2009) Antibiotics in the aquatic environment-a review-part I. Chemosphere 75: 417-434. doi: 10.1016/j.chemosphere.2008.11.086
    [9] Novo A, Andre S, Viana P, et al. (2013) Antibiotic resistance, antimicrobial residues and bacterial community composition in urban wastewater. Water Res 47: 1875-1887. doi: 10.1016/j.watres.2013.01.010
    [10] Margot J, Rossi L, Barry DA, et al. (2015) A review of the fate of micropollutants in wastewater treatment plants. Wiley Interdiscip Rev Water 2: 457-487. doi: 10.1002/wat2.1090
    [11] Sui Q, Cao X, Lu S, et al. (2015) Occurrence, sources and fate of pharmaceuticals and personal care products in the groundwater: A review. Emerg Contam 1: 14-24. doi: 10.1016/j.emcon.2015.07.001
    [12] Du B, Price AE, Scott WC, et al. (2014) Science of the Total Environment Comparison of contaminants of emerging concern removal, discharge, and water quality hazards among centralized and on-site wastewater treatment system ef fl uents receiving common wastewater in fl uent. Sci Total Environ 466-467: 976-984. doi: 10.1016/j.scitotenv.2013.07.126
    [13] Decision 2018/840 (2018) Commission Implementing Decision (EU) 2018/840 of 5 June 2018 establishing a watch list of substances for Union-wide monitoring in the field of water policy pursuant to Directive 2008/105/EC of the European Parliament and of the Council and repealing Comm.
    [14] Directive 2000/60/EC (2000) Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy.
    [15] Directive 2008/105/EC Environmental quality standards applicable to surface water.
    [16] Directive 2013/39/EU Directive 2013/39/EU of the European Parliament and of the Council of 12 August 2013 amending Directives 2000/60/EC and 2008/105/EC as regards priority substances in the field of water policy Text with EEA relevance.
    [17] Decision 2015/495/EU Commission Implementing Decision (EU) 2015/495 of 20 March 2015 establishing a watch list of substances for Union-wide monitoring in the field of water policy pursuant to Directive 2008/105/EC of the European Parliament and of the Council (notified under.
    [18] aus der Beek T, Weber FA, Bergmann A, et al. (2016) Pharmaceuticals in the environment: Global occurrence and potential cooperative action under the Strategic Approach to International Chemicals Management (SAICM). UBA Texte 67/2016.
    [19] Gogoi A, Mazumder P, Kumar V, et al. (2018) Groundwater for Sustainable Development Occurrence and fate of emerging contaminants in water environment : A review. Groundw Sustain Dev 6: 169-180. doi: 10.1016/j.gsd.2017.12.009
    [20] Liu J, Dan X, Lu G, et al. (2018) Investigation of pharmaceutically active compounds in an urban receiving water: Occurrence, fate and environmental risk assessment. Ecotoxicol Environ Saf 154: 214-220. doi: 10.1016/j.ecoenv.2018.02.052
    [21] Zhang R, Zhang R, Yu K, et al. (2018) Occurrence, sources and transport of antibiotics in the surface water of coral reef regions in the South China Sea: Potential risk to coral growth. Environ Pollut 232: 450-457. doi: 10.1016/j.envpol.2017.09.064
    [22] Jelic A, Gros M, Ginebreda A, et al. (2011) Occurrence, partition and removal of pharmaceuticals in sewage water and sludge during wastewater treatment. Water Res 45: 1165-1176. doi: 10.1016/j.watres.2010.11.010
    [23] Biel-Maeso M, Baena-Nogueras RM, Corada-Fernández C, et al. (2018) Occurrence, distribution and environmental risk of pharmaceutically active compounds (PhACs) in coastal and ocean waters from the Gulf of Cadiz (SW Spain). Sci Total Environ 612: 649-659. doi: 10.1016/j.scitotenv.2017.08.279
    [24] Mirzaei R, Yunesian M, Nasseri S, et al. (2018) Occurrence and fate of most prescribed antibiotics in different water environments of Tehran, Iran. Sci Total Environ 619-620: 446-459. doi: 10.1016/j.scitotenv.2017.07.272
    [25] Paredes L, Omil F, Lema JM, et al. (2018) What happens with organic micropollutants during UV disinfection in WWTPs? A global perspective from laboratory to full-scale. J Hazard Mater 342: 670-678.
    [26] Kostich MS, Batt AL, Lazorchak JM (2014) Concentrations of prioritized pharmaceuticals in effluents from 50 large wastewater treatment plants in the US and implications for risk estimation. Environ Pollut 184: 354-359. doi: 10.1016/j.envpol.2013.09.013
    [27] Orias F, Perrodin Y (2013) Characterisation of the ecotoxicity of hospital effluents: A review. Sci Total Environ 454-455: 250-276. doi: 10.1016/j.scitotenv.2013.02.064
    [28] Botero-Coy AM, Martínez-Pachón D, Boix C, et al. (2018) 'An investigation into the occurrence and removal of pharmaceuticals in Colombian wastewater'. Sci Total Environ 642: 842-853. doi: 10.1016/j.scitotenv.2018.06.088
    [29] Wang B, Dai G, Deng S, et al. (2017) Large-scale enzymatic membrane reactors for tetracycline degradation in WWTP effluents. Water Res 45: 1-11.
    [30] Tiedeken EJ, Tahar A, McHugh B, et al. (2017) Monitoring, sources, receptors, and control measures for three European Union watch list substances of emerging concern in receiving waters-A 20 year systematic review. Sci Total Environ 574: 1140-1163. doi: 10.1016/j.scitotenv.2016.09.084
    [31] Oulton RL, Kohn T, Cwiertny DM (2010) Pharmaceuticals and personal care products in effluent matrices: A survey of transformation and removal during wastewater treatment and implications for wastewater management. J Environ Monit 12: 1956. doi: 10.1039/c0em00068j
    [32] Ahmed MB, Zhou JL, Ngo HH, et al. (2016) Progress in the biological and chemical treatment technologies for emerging contaminant removal from wastewater: A critical review. J Hazard Mater 323: 274-298.
    [33] Cecconet D, Molognoni D, Callegari A, et al. (2017) Biological combination processes for efficient removal of pharmaceutically active compounds from wastewater: A review and future perspectives. J Environ Chem Eng 5: 3590-3603. doi: 10.1016/j.jece.2017.07.020
    [34] Bayona JM (2013) Removal of Pharmaceutical Compounds from Wastewater and Surface Water by Natural Treatments. 62.
    [35] Cole S (1998) The Emergence of Treatment Wetlands. Environ Sci Technol 32: 218A-223A. doi: 10.1021/es9834733
    [36] Verlicchi P, Zambello E (2014) How efficient are constructed wetlands in removing pharmaceuticals from untreated and treated urban wastewaters? A review. Sci Total Environ 470-471: 1281-1306. doi: 10.1016/j.scitotenv.2013.10.085
    [37] Fountoulakis MS, Terzakis S, Chatzinotas A, et al. (2009) Pilot-scale comparison of constructed wetlands operated under high hydraulic loading rates and attached biofilm reactors for domestic wastewater treatment. Sci Total Environ 407: 2996-3003. doi: 10.1016/j.scitotenv.2009.01.005
    [38] Hijosa-Valsero M, Fink G, Schlüsener MP, et al. (2011) Removal of antibiotics from urban wastewater by constructed wetland optimization. Chemosphere 83: 713-719. doi: 10.1016/j.chemosphere.2011.02.004
    [39] Hijosa-Valsero M, Matamoros V, Martín-Villacorta J, et al. (2010) Assessment of full-scale natural systems for the removal of PPCPs from wastewater in small communities. Water Res 44: 1429-1439. doi: 10.1016/j.watres.2009.10.032
    [40] Dai Y, Dan A, Yang Y, et al. (2016) Factors Affecting Behavior of Phenolic Endocrine Disruptors, Estrone and Estradiol, in Constructed Wetlands for Domestic Sewage Treatment. EnvironSciTechnol 50: 11844-11852.
    [41] Song H, Nakano K, Taniguchi T, et al. (2009) Estrogen removal from treated municipal effluent in small-scale constructed wetland with different depth. Bioresour Technol 100: 2945-2951. doi: 10.1016/j.biortech.2009.01.045
    [42] Verlicchi P, Galletti A, Petrovic M, et al. (2013) Removal of selected pharmaceuticals from domestic wastewater in an activated sludge system followed by a horizontal subsurface flow bed-Analysis of their respective contributions. Sci Total Environ 454-455: 411-425. doi: 10.1016/j.scitotenv.2013.03.044
    [43] Song H-L, Yang X-L, Nakano K, et al. (2011) Elimination of estrogens and estrogenic activity from sewage treatment works effluents in subsurface and surface flow constructed wetlands. Int J Environ Anal Chem 91: 600-614. doi: 10.1080/03067319.2010.496046
    [44] Hu X, Bao Y, Hu J, et al. (2017) Occurrence of 25 pharmaceuticals in Taihu Lake and their removal from two urban drinking water treatment plants and a constructed wetland. Env Sci Pollut Res 24: 14889-14902. doi: 10.1007/s11356-017-8830-y
    [45] Kimura K, Hara H, Watanabe Y (2007) Elimination of selected acidic pharmaceuticals from municipal wastewater by an activated sludge system and membrane bioreactors. Environ Sci Technol 41: 3708-3714. doi: 10.1021/es061684z
    [46] Fazal S, Zhang B, Zhong Z, et al. (2015) Industrial Wastewater Treatment by Using MBR (Membrane Bioreactor) Review Study. J Environ Prot (Irvine, Calif) 06: 584-598. doi: 10.4236/jep.2015.66053
    [47] Carmosini N, Lee LS (2009) Ciprofloxacin sorption by dissolved organic carbon from reference and bio-waste materials. Chemosphere 77: 813-820. doi: 10.1016/j.chemosphere.2009.08.003
    [48] Zuehlke S, Duennbier U, Lesjean B, et al. (2006) Long-Term Comparison of Trace Organics Removal Performances Between Conventional and Membrane Activated Sludge Processes. Water Environ Res 78: 2480-2486. doi: 10.2175/106143006X111826
    [49] Baumgarten S (2007) Membranbioreaktoren zur industriellen Abwasserreinigung.
    [50] Janssens R, Mandal MK, Dubey KK, et al. (2017) Slurry photocatalytic membrane reactor technology for removal of pharmaceutical compounds from wastewater: Towards cytostatic drug elimination. Sci Total Environ 599-600: 612-626. doi: 10.1016/j.scitotenv.2017.03.253
    [51] Abass OK, Wu X, Guo Y, et al. (2015) Membrane Bioreactor in China: A Critical Review. Int J Membr Sci Technol 2: 29-47. doi: 10.15379/2410-1869.2015.02.02.04
    [52] Besha AT, Gebreyohannes AY, Tufa RA, et al. (2017) Removal of emerging micropollutants by activated sludge process and membrane bioreactors and the effects of micropollutants on membrane fouling: A review. J Environ Chem Eng 5: 2395-2414. doi: 10.1016/j.jece.2017.04.027
    [53] Fan H, Li J, Zhang L, et al. (2014) Contribution of sludge adsorption and biodegradation to the removal of five pharmaceuticals in a submerged membrane bioreactor. Biochem Eng J 88: 101-107. doi: 10.1016/j.bej.2014.04.008
    [54] Kovalova L, Siegrist H, Singer H, et al. (2012) Hospital wastewater treatment by membrane bioreactor: Performance and efficiency for organic micropollutant elimination. Environ Sci Technol 46: 1536-1545. doi: 10.1021/es203495d
    [55] Le T, Ng C, Tran NH, et al. (2018) Removal of antibiotic residues, antibiotic resistant bacteria and antibiotic resistance genes in municipal wastewater by membrane bioreactor systems. Water Res 498-508.
    [56] Zhang W, Ding L, Luo J, et al. (2016) Membrane fouling in photocatalytic membrane reactors (PMRs) for water and wastewater treatment: A critical review. Chem Eng J 302: 446-458. doi: 10.1016/j.cej.2016.05.071
    [57] Huang BC, Guan YF, Chen W, et al. (2017) Membrane fouling characteristics and mitigation in a coagulation-assisted microfiltration process for municipal wastewater pretreatment. Water Res 123: 216-223. doi: 10.1016/j.watres.2017.06.080
    [58] Liébana R, Arregui L, Belda I, et al. (2015) Membrane bioreactor wastewater treatment plants reveal diverse yeast and protist communities of potential significance in biofouling. Biofouling 31: 71-82. doi: 10.1080/08927014.2014.998206
    [59] Oppenländer T (2003) Photochemical Purification of Water and Air: Advanced Oxidation Processes (AOPs): Principles, Reaction Mechanisms, Reactor Concepts (Chemistry), Weinheim, WILEY-VCH Verlag.
    [60] Parsons S (2004) Advanced Oxidation Processes for Water and Wastewater Treatment, London, IWA Publishing.
    [61] Andreozzi R, Caprio V, Insola A, et al. (1999) Advanced oxidation processes (AOP) for water purification and recovery. Catal Today 53: 51-59. doi: 10.1016/S0920-5861(99)00102-9
    [62] Voigt M, Bartels I, Nickisch-hartfiel A, et al. (2018) Elimination of macrolides in water bodies using photochemical oxidation. AIMS Environ Sci 5: 372-388. doi: 10.3934/environsci.2018.5.372
    [63] Homem V, Santos L (2011) Degradation and removal methods of antibiotics from aqueous matrices--a review. J Environ Manage 92: 2304-2347. doi: 10.1016/j.jenvman.2011.05.023
    [64] Monteagudo JM, Durán A, San Martín I (2014) Mineralization of wastewater from the pharmaceutical industry containing chloride ions by UV photolysis of H2O2/Fe(II) and ultrasonic irradiation. J Environ Manage 141: 61-69. doi: 10.1016/j.jenvman.2014.03.020
    [65] Shahidi D, Roy R, Azzouz A (2015) Advances in catalytic oxidation of organic pollutants-Prospects for thorough mineralization by natural clay catalysts. Appl Catal B Environ 174-175: 277-292. doi: 10.1016/j.apcatb.2015.02.042
    [66] Brillas E (2014) A review on the degradation of organic pollutants in waters by UV photoelectro-fenton and solar photoelectro-fenton. J Braz Chem Soc 25: 393-417.
    [67] Voigt M, Bartels I, Nickisch-Hartfiel A, et al. (2017) Photoinduced degradation of sulfonamides, kinetic, and structural characterization of transformation products and assessment of environmental toxicity. Toxicol Environ Chem 99: 1304-1327. doi: 10.1080/02772248.2017.1373777
    [68] Jelic a., Michael I, Achilleos a., et al. (2013) Transformation products and reaction pathways of carbamazepine during photocatalytic and sonophotocatalytic treatment. J Hazard Mater 263: 177-186. doi: 10.1016/j.jhazmat.2013.07.068
    [69] Vasconcelos TG, Henriques DM, König A, et al. (2009) Photo-degradation of the antimicrobial ciprofloxacin at high pH: Identification and biodegradability assessment of the primary by-products. Chemosphere 76: 487-493. doi: 10.1016/j.chemosphere.2009.03.022
    [70] Suslick S, Fang M (1999) Acoustic cavitation and its chemical consequences. Phil Trans R Soc Lond A 335-353.
    [71] Torres-Palma RA, Serna-Galvis EA (2018) Sonolysis, Advanced Oxidation Processes for Waste Water Treatment, Elsevier, 177-213.
    [72] Serna-Galvis EA, Botero-Coy AM, Martínez-Pachón D, et al. (2019) Degradation of seventeen contaminants of emerging concern in municipal wastewater effluents by sonochemical advanced oxidation processes. Water Res 154: 349-360. doi: 10.1016/j.watres.2019.01.045
    [73] Zoschke K, Börnick H, Worch E (2014) Vacuum-UV radiation at 185 nm in water treatment-a review. Water Res 52: 131-145. doi: 10.1016/j.watres.2013.12.034
    [74] Crapulli F, Santoro D, Sasges MR, et al. (2014) Mechanistic modeling of vacuum UV advanced oxidation process in an annular photoreactor. Water Res 64: 209-225. doi: 10.1016/j.watres.2014.06.048
    [75] Heit G, Neuner A, Saugy P, et al. (1998) Vacuum-UV (172 nm) Actinometry. The Quantum Yield of the Photolysis of Water. J Phys Chem A 5639: 5551-5561.
    [76] Ratpukdi T (2014) Degradation of Paracetamol and Norfloxacin in Aqueous Solution Using Vacuum Ultraviolet (VUV) Process. J Clean Energy Technol 2: 168-170.
    [77] Szabó RK, Megyeri C, Illés E, et al. (2011) Phototransformation of ibuprofen and ketoprofen in aqueous solutions. Chemosphere 84: 1658-1663. doi: 10.1016/j.chemosphere.2011.05.012
    [78] Voigt M, Jaeger M (2017) On the photodegradation of azithromycin, erythromycin and tylosin and their transformation products-A kinetic study. Sustain Chem Pharm 5: 131-140. doi: 10.1016/j.scp.2016.12.001
    [79] Gottschalk C, Libra JA, Saupe A (2010) Ozonation of Water and Waste Water-A Practical Guide to Understanding Ozone and its Applications, Weinheim, WILEY-VCH Verlag GmbH & Co. KGaA.
    [80] Antoniou MG, Hey G, Rodríguez Vega S, et al. (2013) Required ozone doses for removing pharmaceuticals from wastewater effluents. Sci Total Environ 456-457: 42-49. doi: 10.1016/j.scitotenv.2013.03.072
    [81] Ribeiro AR, Nunes OC, Pereira MFR, et al. (2015) An overview on the advanced oxidation processes applied for the treatment of water pollutants defined in the recently launched Directive 2013/39/EU. Environ Int 75: 33-51. doi: 10.1016/j.envint.2014.10.027
    [82] Matilainen A, Sillanpää M (2010) Removal of natural organic matter from drinking water by advanced oxidation processes. Chemosphere 80: 351-365. doi: 10.1016/j.chemosphere.2010.04.067
    [83] Tong L, Eichhorn P, Pérez S, et al. (2011) Photodegradation of azithromycin in various aqueous systems under simulated and natural solar radiation: kinetics and identification of photoproducts. Chemosphere 83: 340-348. doi: 10.1016/j.chemosphere.2010.12.025
    [84] Frontistis Z, Kouramanos M, Moraitis S, et al. (2015) UV and simulated solar photodegradation of 17α-ethynylestradiol in secondary-treated wastewater by hydrogen peroxide or iron addition. Catal Today 252: 84-92. doi: 10.1016/j.cattod.2014.10.012
    [85] Ma X, Zhang C, Deng J, et al. (2015) Simultaneous degradation of estrone, 17β-estradiol and 17α-ethinyl estradiol in an aqueous UV/H2o2 system. Int J Environ Res Public Health 12: 12016-12029. doi: 10.3390/ijerph121012016
    [86] Voigt M, Savelsberg C, Jaeger M (2017) Photodegradation of the antibiotic spiramycin studied by high-performance liquid quadrupole time-of-flight mass spectrometry. Toxicol Environ Chem 99: 624-640. doi: 10.1080/02772248.2017.1280039
    [87] Hernández F, Bakker J, Bijlsma L, et al. (2019) The role of analytical chemistry in exposure science: Focus on the aquatic environment. Chemosphere 564-583.
    [88] Voigt M, Savelsberg C, Jaeger M (2018) Identification of Pharmaceuticals in The Aquatic Environment Using HPLC-ESI-Q-TOF-MS and Elimination of Erythromycin Through Photo-Induced Degradation. J Vis Exp.
    [89] Bobu M, Yediler A, Siminiceanu I, et al. (2013) Comparison of different advanced oxidation processes for the degradation of two fluoroquinolone antibiotics in aqueous solutions. J Environ Sci Heal Part A 48: 251-262.
    [90] Yu F, Li Y, Han S, et al. (2016) Adsorptive removal of antibiotics from aqueous solution using carbon materials. Chemosphere 153: 365-385. doi: 10.1016/j.chemosphere.2016.03.083
    [91] Sharif F, Westerhoff P, Herckes P (2013) Sorption of trace organics and engineered nanomaterials onto wetland plant material. Environ Sci Process Impacts 15: 267-274. doi: 10.1039/C2EM30613A
    [92] Gao Y, Li Y, Zhang L, et al. (2012) Adsorption and removal of tetracycline antibiotics from aqueous solution by graphene oxide. J Colloid Interface Sci 368: 540-546. doi: 10.1016/j.jcis.2011.11.015
    [93] Mailler R, Gasperi J, Coquet Y, et al. (2016) Removal of emerging micropollutants from wastewater by activated carbon adsorption: Experimental study of different activated carbons and factors influencing the adsorption of micropollutants in wastewater. J Environ Chem Eng 4: 1102-1109. doi: 10.1016/j.jece.2016.01.018
    [94] Kaub M, Biebersdorf N (2014) Kläranlage Höxter 4. Reinigungsstufe zur Elimination von Mikroschadstoffen, Bochum.
    [95] Grover DP, Zhou JL, Frickers PE, et al. (2011) Improved removal of estrogenic and pharmaceutical compounds in sewage effluent by full scale granular activated carbon : Impact on receiving river water. J Hazard Mater 185: 1005-1011. doi: 10.1016/j.jhazmat.2010.10.005
    [96] Kovalova L, Siegrist H, Von Gunten U, et al. (2013) Elimination of micropollutants during post-treatment of hospital wastewater with powdered activated carbon, ozone, and UV. Environ Sci Technol 47: 7899-7908. doi: 10.1021/es400708w
    [97] Mailler R, Gasperi J, Coquet Y, et al. (2016) Removal of a wide range of emerging pollutants from wastewater treatment plant discharges by micro-grain activated carbon in fluidized bed as tertiary treatment at large pilot scale. Sci Total Environ 542: 983-996. doi: 10.1016/j.scitotenv.2015.10.153
    [98] Lima DRS, Baêta BEL, Aquino SF, et al. (2014) Removal of Pharmaceuticals and Endocrine Disruptor Compounds from Natural Waters by Clarification Associated with Powdered Activated Carbon. Water Air Soil Pollut 225.
    [99] Rubirola A, Llorca M, Rodriguez-Mozaz S, et al. (2014) Characterization of metoprolol biodegradation and its transformation products generated in activated sludge batch experiments and in full scale WWTPs. Water Res 63: 21-32. doi: 10.1016/j.watres.2014.05.031
    [100] Kårelid V, Larsson G, Björlenius B (2017) Pilot-scale removal of pharmaceuticals in municipal wastewater: Comparison of granular and powdered activated carbon treatment at three wastewater treatment plants. J Environ Manage 193: 491-502. doi: 10.1016/j.jenvman.2017.02.042
    [101] Li L, Quinlivan PA, Knappe DRU (2002) Effects of activated carbon surface chemistry and pore structure on the adsorption of organic contaminants from aqueous solution. Carbon N Y 40: 2085-2100. doi: 10.1016/S0008-6223(02)00069-6
    [102] Benstoem F, Nahrstedt A, Boehler M, et al. (2017) Performance of granular activated carbon to remove micropollutants from municipal wastewater-A meta-analysis of pilot- and large-scale studies. Chemosphere 185: 105-118. doi: 10.1016/j.chemosphere.2017.06.118
    [103] Margot J, Magnet A (2011) Elimination des micropolluants dans les eaux usées-Essais pilotes à la station d épuration de Lausanne. gwa 7: 487-493.
    [104] Margot J, Kienle C, Magnet A, et al. (2013) Treatment of micropollutants in municipal wastewater: Ozone or powdered activated carbon? Sci Total Environ 461-462: 480-498. doi: 10.1016/j.scitotenv.2013.05.034
    [105] Baresel C, Malmborg J, Ek M, et al. (2016) Removal of pharmaceutical residues using ozonation as intermediate process step at Linköping WWTP, Sweden. Water Sci Technol 73: 2017-2024. doi: 10.2166/wst.2016.045
    [106] Mamo J, Insa S, Monclús H, et al. (2016) Fate of NDMA precursors through an MBR-NF pilot plant for urban wastewater reclamation and the effect of changing aeration conditions. Water Res 102: 383-393. doi: 10.1016/j.watres.2016.06.057
    [107] Hofman-Caris CHM, Siegers WG, van de Merlen K, et al. (2017) Removal of pharmaceuticals from WWTP effluent: Removal of EfOM followed by advanced oxidation. Chem Eng J 327: 514-521. doi: 10.1016/j.cej.2017.06.154
    [108] Cédat B, de Brauer C, Métivier H, et al. (2016) Are UV photolysis and UV/H 2 O 2 process efficient to treat estrogens in waters? Chemical and biological assessment at pilot scale. Water Res 100: 357-366.
    [109] El-taliawy H, Ekblad M, Nilsson F, et al. (2017) Ozonation efficiency in removing organic micro pollutants from wastewater with respect to hydraulic loading rates and different wastewaters. Chem Eng J 325: 310-321. doi: 10.1016/j.cej.2017.05.019
    [110] Ibáñez M, Borova V, Boix C, et al. (2017) UHPLC-QTOF MS screening of pharmaceuticals and their metabolites in treated wastewater samples from Athens. J Hazard Mater 323: 26-35. doi: 10.1016/j.jhazmat.2016.03.078
  • This article has been cited by:

    1. Mohammad Mahdi Azizi, Setareh Abhari, Hedieh Sajedi, Alan Marmorstein, Stitched vision transformer for age-related macular degeneration detection using retinal optical coherence tomography images, 2024, 19, 1932-6203, e0304943, 10.1371/journal.pone.0304943
    2. S. Vishnu Priyan, R. Vinod Kumar, C. Moorthy, V.S. Nishok, A fusion of deep neural networks and game theory for retinal disease diagnosis with OCT images, 2024, 32, 08953996, 1011, 10.3233/XST-240027
    3. P. T. Karule, Sujata B. Bhele, Prasanna Palsodkar, Poonam T. Agarkar, Hirendra R. Hajare, Prashant R. Patil, 2024, Detection of Multi-Class Multi-Label Ophthalmological Diseases in Retinal Fundus Images Using Machine Learning, 979-8-3503-1901-9, 1, 10.1109/ICICET59348.2024.10616291
    4. Yash Mori, Nandini Modi, 2024, Enhancing OCT Image Classification for Retinal Disease Diagnosis: A Novel Approach Using Squeeze-and-Excitation ResNet, 979-8-3503-8459-8, 940, 10.1109/AIC61668.2024.10730829
    5. Muhammed Enes Atik, İbrahim Kocak, Nihat Sayin, Sadik Etka Bayramoglu, Ahmet Ozyigit, Integration of Optical Coherence Tomography Images and Real‐Life Clinical Data for Deep Learning Modeling: A Unified Approach in Prognostication of Diabetic Macular Edema, 2024, 1864-063X, 10.1002/jbio.202400315
    6. G. Jeyasri, R. Karthiyayini, Deep learning based retinal disease classification using an autoencoder and generative adversarial network, 2025, 108, 17468094, 107852, 10.1016/j.bspc.2025.107852
    7. Pavithra Mani, Neelaveni Ramachandran, Palanichamy Naveen, Prasanna Venkatesh Ramesh, An enhanced lightweight transformer-based framework for accurate retinal disease classification from OCT images, 2025, 0972-8821, 10.1007/s12596-025-02793-6
  • Reader Comments
  • © 2020 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(12118) PDF downloads(720) Cited by(20)

Figures and Tables

Figures(13)  /  Tables(8)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog