An amalgamation of deep neural networks optimized with Salp swarm algorithm for cervical cancer detection

Computers & Electrical Engineering, Journal Year: 2025, Volume and Issue: 123, P. 110106 - 110106

Published: Jan. 28, 2025

Language: Английский

---

Deep Learning Based Segmentation of Magnetic Resonance Cardiac Images

Lecture notes in electrical engineering, Journal Year: 2025, Volume and Issue: unknown, P. 210 - 218

Published: Jan. 1, 2025

Language: Английский

---

BrainTumNet: multi-task deep learning framework for brain tumor segmentation and classification using adaptive masked transformers

Frontiers in Oncology, Journal Year: 2025, Volume and Issue: 15

Published: May 20, 2025

Background and objective Accurate diagnosis of brain tumors significantly impacts patient prognosis treatment planning. Traditional diagnostic methods primarily rely on clinicians’ subjective interpretation medical images, which is heavily dependent physician experience limited by time consumption, fatigue, inconsistent diagnoses. Recently, deep learning technologies, particularly Convolutional Neural Networks (CNN), have achieved breakthrough advances in image analysis, offering a new paradigm for automated precise diagnosis. However, existing research largely focuses single-task modeling, lacking comprehensive solutions that integrate tumor segmentation with classification This study aims to develop multi-task model type classification. Methods The included 485 pathologically confirmed cases, comprising T1-enhanced MRI sequence images high-grade gliomas, metastatic tumors, meningiomas. dataset was proportionally divided into training (378 cases), testing (109 external validation (51 cases) sets. We designed implemented BrainTumNet, learning-based framework featuring an improved encoder-decoder architecture, adaptive masked Transformer, multi-scale feature fusion strategy simultaneously perform region pathological Five-fold cross-validation employed result verification. Results In the test set evaluation, BrainTumNet Intersection over Union (IoU) 0.921, Hausdorff Distance (HD) 12.13, Dice Similarity Coefficient (DSC) 0.91 segmentation. For classification, it attained accuracy 93.4% Area Under ROC Curve (AUC) 0.96. Performance remained stable set, confirming model’s generalization capability. Conclusion proposed achieves high-precision through strategy. Experimental results demonstrate strong potential clinical application, providing reliable auxiliary information preoperative assessment decision-making cases.

Language: Английский

---

An explainable Liquid Neural Network combined with path aggregation residual network for an accurate brain tumor diagnosis

Computers & Electrical Engineering, Journal Year: 2024, Volume and Issue: 122, P. 109999 - 109999

Published: Dec. 18, 2024

Language: Английский

---