Multi-Class Brain Malignant Tumor Diagnosis in Magnetic Resonance Imaging Using Convolutional Neural Networks

Brain Research Bulletin, Journal Year: 2025, Volume and Issue: unknown, P. 111329 - 111329

Published: April 1, 2025

To reduce the clinical misdiagnosis rate of glioblastoma (GBM), primary central nervous system lymphoma (PCNSL), and brain metastases (BM), which are common malignant tumors with similar radiological features, we propose a new CNN-based model, FoTNet. The model integrates frequency-based channel attention layer Focal Loss to address class imbalance issue caused by limited data available for PCNSL. A multi-center MRI dataset was constructed collecting integrating from Zhejiang University School Medicine's Sir Run Shaw Hospital, along public datasets UPENN TCGA. includes T1-weighted contrast-enhanced (T1-CE) images 58 GBM, 82 PCNSL, 269 BM cases, were divided into training testing sets in 5:2 ratio. FoTNet achieved classification accuracy 92.5% an average AUC 0.9754 on test set, significantly outperforming existing machine learning deep methods distinguishing between BM. Through multiple validations, has proven be effective robust tool accurately classifying these tumors, providing strong support preoperative diagnosis assisting clinicians making more informed treatment decisions.

Language: Английский

Differential evolution-driven optimized ensemble network for brain tumor detection

International Journal of Machine Learning and Cybernetics, Journal Year: 2025, Volume and Issue: unknown

Published: April 16, 2025

Language: Английский

Enhancing PM2.5 Air Pollution Prediction Performance by Optimizing the Echo State Network (ESN) Deep Learning Model Using New Metaheuristic Algorithms

Urban Science, Journal Year: 2025, Volume and Issue: 9(5), P. 138 - 138

Published: April 23, 2025

Air pollution presents significant risks to both human health and the environment. This study uses air meteorological data develop an effective deep learning model for hourly PM2.5 concentration predictions in Tehran, Iran. evaluates efficient metaheuristic algorithms optimizing hyperparameters improve accuracy of predictions. The optimal feature set was selected using Variance Inflation Factor (VIF) Boruta-XGBoost methods, which indicated elimination NO, NO2, NOx. highlighted PM10 as most important feature. Wavelet transform then applied extract 40 features enhance prediction accuracy. Hyperparameters weights matrices Echo State Network (ESN) were determined algorithms, with Salp Swarm Algorithm (SSA) demonstrating superior performance. evaluation different criteria revealed that ESN-SSA outperformed other hybrids original ESN, LSTM, GRU models.

Language: Английский