An integrated approach based on Gaussian noises-based data augmentation method and AdaBoost model to predict faecal coliforms in rivers with small dataset

Bilali, Ali El; Taleb, Abdeslam; Bahlaoui, M. Abdellah; Brouziyne, Youssef

doi:10.1016/j.jhydrol.2021.126510

Cited by 46 publications

(12 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the pre-trained deep CNN models, four metrics are employed for verifying the quality of the COVID-19 classification results, including accuracy, sensitivity, specificity, and F-Measure criteria [ 20 , 21 ]. We show the quantitative comparison of different augmentation strategies in Fig.…”

Section: Resultsmentioning

confidence: 99%

“…Typically, noise-based data augmentation in DL is performed when there is a possibility of image data being corrupted by noise [ 16 ]. Data augmentation by noise addition is a strategy that improves the robustness and generalization of CNNs [ [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] ]. Moreno-Barea et al [ 21 ] tested the noise injection to images from a Gaussian distribution, and showed it to be useful for improving CNN-based classification performance [ 23 , 24 ].…”

Section: Introductionmentioning

confidence: 99%

“…Data augmentation by noise addition is a strategy that improves the robustness and generalization of CNNs [ [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] ]. Moreno-Barea et al [ 21 ] tested the noise injection to images from a Gaussian distribution, and showed it to be useful for improving CNN-based classification performance [ 23 , 24 ]. Sezer and Sezer [ 25 ] proposed a data augmentation approach, where CNN-based speckle-noise reduction is used in the neonatal hip ultrasound image classification task.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning-to-augment strategy using noisy and denoised data: Improving generalizability of deep CNN for the detection of COVID-19 in X-ray images

Momeny

Neshat

Hussain

et al. 2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

Chest X-ray images are used in deep convolutional neural networks for the detection of COVID-19, the greatest human challenge of the 21st century. Robustness to noise and improvement of generalization are the major challenges in designing these networks. In this paper, we introduce a strategy for data augmentation using the determination of the type and value of noise density to improve the robustness and generalization of deep CNNs for COVID-19 detection. Firstly, we present a learning-to-augment approach that generates new noisy variants of the original image data with optimized noise density. We apply a Bayesian optimization technique to control and choose the optimal noise type and its parameters. Secondly, we propose a novel data augmentation strategy, based on denoised X-ray images, that uses the distance between denoised and original pixels to generate new data. We develop an autoencoder model to create new data using denoised images corrupted by the Gaussian and impulse noise. A database of chest X-ray images, containing COVID-19 positive, healthy, and non-COVID pneumonia cases, is used to fine-tune the pre-trained networks (AlexNet, ShuffleNet, ResNet18, and GoogleNet). The proposed method performs better results compared to the state-of-the-art learning to augment strategies in terms of sensitivity (0.808), specificity (0.915), and F-Measure (0.737). The source code of the proposed method is available at https://github.com/mohamadmomeny/Learning-to-augment-strategy .

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Learning-to-augment strategy using noisy and denoised data: Improving generalizability of deep CNN for the detection of COVID-19 in X-ray images

Momeny

Neshat

Hussain

et al. 2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

show abstract

“…From the data pretreatment aspect, the following strategies can be considered to offset the inherent shortage of the small sample size. First, Bayesian approaches can be used to quantify the data uncertainty, incorporate prior knowledge, estimate parameters, and tune hyperparameters. − Second, application of data augmentation may enlarge the original dataset for the model, including exploratory data analysis, connecting with original data, selecting important input features, generating virtual samples, and iterating with multiple subsamples. , Also, a specific algorithm can be used for a limited number of samples. Nam et al proposed a learning to self-generate diverse few-shot tasks by treating randomly chosen columns as a target label .…”

Section: Discussionmentioning

confidence: 99%

Investigating Biodegradation of 1,4-Dioxane by Groundwater and Soil Microbiomes: Insights into Microbial Ecology and Process Prediction

Miao,

Zhou,

Zheng

et al. 2023

ACS EST Water

View full text Add to dashboard Cite

Although microorganisms play significant roles in bioremediation, their contributions to long-term site characteristics during and after active treatment need to be fully elucidated. This study described microbial ecology dynamics in 1,4-dioxane- and chlorinated solvents-contaminated groundwater in laboratory microcosms. Bioaugmented Pseudonocardia dioxanivorans CB1190 improved 1,4-dioxane removal, with increased carbohydrate and amino acid metabolism, but was eventually outcompeted by native microbes. The original microbiomes were perturbed and divergent but tended to be similar over time. Dechlorinating bacteria co-existed in the same niche, whereas CB1190 had more negative interactions in the shared niche. Multiple regression and classification machine learning models were built by using microbial taxa to predict the degradation process; the ensemble regression model provided most accurate prediction of 1,4-dioxane concentrations (R 2 = 0.81 ± 0.17). Among the classification models, the support vector machine performed the best in differentiating the contamination levels (accuracy at 0.67 ± 0.07, kappa at 0.56 ± 0.10). The ensemble model predicted the 1,4-dioxane concentrations and relative duration of contamination with independent microbial datasets from a field study, and the results aligned with the geographic and hydrological information from monitoring wells. This study introduces the application of machine learning in microbiome-based diagnostics for groundwater remediation and evaluation, providing valuable methods for future research and practice.

show abstract

“…Adaptive boosting (AdaBoost) is a tree-based ensemble ML model [39,40]. Recently, this approach appeared to be an efficient regression model in environmental sciences, namely, for regression and data-based augmentation techniques [15,41].…”

Section: Adaboostmentioning

confidence: 99%

Predicting Daily Suspended Sediment Load Using Machine Learning and NARX Hydro-Climatic Inputs in Semi-Arid Environment

Ezzaouini

Mahé

Kacimi

et al. 2022

Water

Self Cite

View full text Add to dashboard Cite

Sediment transport in basins disturbs the ecological systems of the water bodies and leads to reservoir siltation. Its evaluation is crucial for managing water resources. The practical application of the process-based model can confront some limitations noticed in the lower accuracy during the validation process due to the lack of reliable physical datasets. In this study, we attempt to apply machine-learning-based modeling (ML) to predict the suspended sediment load, using hydro-climatic data as input variables in the semi-arid Bouregreg basin, Morocco. To that end, data for the years 2016 to 2020 were used for the training process, and the validation was performed with 2021 data. The results showed that most ML models have good accuracy, with a Nash–Schiff efficiency (NSE) ranging from 0.47 to 0.80 during the validation phase, which indicates satisfactory performances in predicting the SSL. Furthermore, the models were ranked against their generalization ability (GA), which revealed that the developed models are good to excellent in terms of GA. Overall, the present study provides new insight into predicting the SSL in a semi-arid environment, such as the Bouregreg basin.

show abstract

An integrated approach based on Gaussian noises-based data augmentation method and AdaBoost model to predict faecal coliforms in rivers with small dataset

Cited by 46 publications

References 61 publications

Learning-to-augment strategy using noisy and denoised data: Improving generalizability of deep CNN for the detection of COVID-19 in X-ray images

Learning-to-augment strategy using noisy and denoised data: Improving generalizability of deep CNN for the detection of COVID-19 in X-ray images

Investigating Biodegradation of 1,4-Dioxane by Groundwater and Soil Microbiomes: Insights into Microbial Ecology and Process Prediction

Predicting Daily Suspended Sediment Load Using Machine Learning and NARX Hydro-Climatic Inputs in Semi-Arid Environment

Contact Info

Product

Resources

About