Computers & Electrical Engineering, Год журнала: 2025, Номер 126, С. 110412 - 110412
Опубликована: Июнь 3, 2025
Язык: Английский
Results in Engineering, Год журнала: 2025, Номер unknown, С. 104349 - 104349
Опубликована: Фев. 1, 2025
Язык: Английский
Процитировано
3
Neural Computing and Applications, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 18, 2024
Язык: Английский
Процитировано
13
Computers & Electrical Engineering, Год журнала: 2024, Номер 120, С. 109795 - 109795
Опубликована: Окт. 20, 2024
Язык: Английский
Процитировано
9
Journal of the Brazilian Society of Mechanical Sciences and Engineering, Год журнала: 2025, Номер 47(4)
Опубликована: Март 5, 2025
Процитировано
1
Frontiers in Oncology, Год журнала: 2025, Номер 15
Опубликована: Апрель 9, 2025
Glioblastoma (GBM), the most aggressive primary brain tumor, poses a significant challenge in predicting patient survival due to its heterogeneity and resistance treatment. Accurate prediction is essential for optimizing treatment strategies improving clinical outcomes. This study utilized metadata from 135 GBM patients, including demographic, clinical, molecular variables such as age, Karnofsky Performance Status (KPS), MGMT promoter methylation, EGFR amplification. Six machine learning models-XGBoost, Random Forests, Support Vector Machines, Artificial Neural Networks, Extra Trees Regressor, K- Nearest Neighbors-were employed classify patients into predefined categories. Data preprocessing included label encoding categorical MinMax scaling numerical features. Model performance was assessed using ROC-AUC accuracy metrics, with hyperparameters optimized through grid search. XGBoost demonstrated highest predictive accuracy, achieving mean of 0.90 an 0.78. Ensemble models outperformed simpler classifiers, emphasizing value metadata. The identified key prognostic markers, methylation KPS, contributors prediction. application offers robust approach survival. highlights potential ML enhance decision-making contribute personalized strategies, focus on reliability, interpretability.
Язык: Английский
Процитировано
1
Euro-Mediterranean Journal for Environmental Integration, Год журнала: 2024, Номер unknown
Опубликована: Окт. 3, 2024
Abstract Membrane filtration processes have demonstrated remarkable effectiveness in wastewater treatment, achieving high contaminant removal and producing high-quality effluent suitable for safe reuse. technologies play a primary role combating water scarcity pollution challenges. However, the need more effective strategies to mitigate membrane fouling remains critical concern. Artificial intelligence (AI) modeling offers promising solution by enabling accurate predictions of fouling, thus supporting advanced mitigation strategies. This review examines recent progress application AI models, with particular focus on artificial neural networks (ANNs), simulating treatment processes. It highlights substantial potential ANNs, particularly widely studied multi-layer perceptron (MLP) other emerging configurations, accurately predict thereby enhancing process optimization efforts. The discusses both benefits current limitations AI-based strategies, analyzing studies offer valuable insights designing ANNs capable providing predictions. Specifically, it provides guidance selecting appropriate model architectures, input/output variables, activation functions, training algorithms. Finally, this connect research findings practical applications full-scale plants. Key steps crucial address challenge been identified, emphasizing revolutionize control drive paradigm shift toward efficient sustainable membrane-based treatment.
Язык: Английский
Процитировано
5
Current Opinion in Biomedical Engineering, Год журнала: 2024, Номер 33, С. 100567 - 100567
Опубликована: Ноя. 14, 2024
Язык: Английский
Процитировано
3
Frontiers in Toxicology, Год журнала: 2025, Номер 6
Опубликована: Янв. 8, 2025
The adoption of in vivo digital measures pharmaceutical research and development (R&D) presents an opportunity to enhance the efficiency effectiveness discovering developing new therapeutics. For clinical measures, Digital Medicine Society's (DiMe) V3 Framework is a comprehensive validation framework that encompasses verification, analytical validation, validation. This manuscript describes collaborative efforts adapt this ensure reliability relevance for preclinical context. Verification ensures technologies accurately capture store raw data. Analytical assesses precision accuracy algorithms transform data into meaningful biological metrics. Clinical confirms these reflect or functional states animal models relevant their context use. By widely adopting structured approach, stakeholders-including researchers, technology developers, regulators-can applicability research, ultimately supporting more robust translatable drug discovery processes.
Язык: Английский
Процитировано
0
Foods, Год журнала: 2025, Номер 14(2), С. 247 - 247
Опубликована: Янв. 14, 2025
This review explores the application of convolutional neural networks (CNNs) and recurrent (RNNs) in food safety detection risk prediction. paper highlights advantages CNNs image processing feature recognition, as well powerful capabilities RNNs (especially their variant LSTM) time series data modeling. also makes a comparative analysis many aspects: Firstly, disadvantages traditional prediction methods are compared with deep learning technologies such RNNs. Secondly, similarities differences between fully connected analyzed. Furthermore, statistical modeling discussed. Finally, directions compared. discusses combining these models Internet Things (IoT), blockchain, federated to improve accuracy efficiency warning. this mentions limitations field safety, challenges interpretability model, suggests use interpretable artificial intelligence (XAI) technology transparency model.
Язык: Английский
Процитировано
0
Remote Sensing, Год журнала: 2025, Номер 17(3), С. 460 - 460
Опубликована: Янв. 29, 2025
Atmospheric correction is essential in remote sensing, as it reduces the effects of light absorption and scattering by suspended particles gases, enabling accurate surface reflectance computation from observed Top-of-Atmosphere (TOA) reflectance. Each satellite sensor requires a customized atmospheric processor due to its unique system characteristics. Currently, PeruSAT-1, first Peruvian sensing satellite, does not include this capability image processing pipeline, which poses challenges for effectiveness defense, security, natural disaster management applications. This research investigated pixel-based machine learning methods using Sentinel-2 harmonized Bottom-of-Atmosphere (BOA) images benchmark, alongside additional atmospheric, terrain, acquisition parameters. A robust dataset was developed align data across temporal, spatial, geometric, contextual conditions. Experimental results showed R2 values between 0.886 0.938, RMSE ranging 0.009 0.025 compared benchmarks. Among models tested, Feedforward Neural Network (FFNN) Leaky ReLU activation function achieved best overall performance. These findings confirm robustness approach, offering scalable methodology satellites with similar characteristics establishing foundation pipeline PeruSAT-1. Future work will focus on diversifying spectral seasonal conditions optimizing modeling address shorter wavelengths high-reflectance surfaces.
Язык: Английский
Процитировано
0