Salience Feature Guided Decoupling Network for UAV Forests Flame Detection

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126414 - 126414

Published: Jan. 1, 2025

Language: Английский

---

Automating Prostate Cancer Grading: A Novel Deep Learning Framework for Automatic Prostate Cancer Grade Assessment using Classification and Segmentation

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 6, 2025

Prostate Cancer (PCa) is the second most common cancer in men and affects more than a million people each year. Grading prostate based on Gleason grading system, subjective labor-intensive method for evaluating tissue samples. The variability diagnostic approaches underscores urgent need reliable methods. By integrating deep learning technologies developing automated systems, precision can be improved, human error minimized. present work introduces three-stage framework-based innovative deep-learning system assessing PCa severity using PANDA challenge dataset. After meticulous selection process, 2699 usable cases were narrowed down from initial 5160 after extensive data cleaning. There are three stages proposed framework: classification of grades neural networks (DNNs), segmentation grades, computation International Society Urological Pathology (ISUP) machine classifiers. Four classes patches classified segmented (benign, 3, 4, 5). Patch sampling at different sizes (500 × 500 1000 pixels) was used to optimize processes. performance network enhanced by Self-organized operational (Self-ONN) DeepLabV3 architecture. Based these predictions, distribution percentages grade within whole slide images (WSI) calculated. These features then concatenated into classifiers predict final ISUP grade. EfficientNet_b0 achieved highest F1-score 83.83% classification, while + architecture self-ONN EfficientNet encoder Dice Similarity Coefficient (DSC) score 84.9% segmentation. Using RandomForest (RF) classifier, framework quadratic weighted kappa (QWK) 0.9215. Deep frameworks being developed automatically have shown promising results. In addition, it provides prospective approach prognostic tool that produce clinically significant results efficiently reliably. Further investigations needed evaluate framework's adaptability effectiveness across various clinical scenarios.

Language: Английский

---

Detection and segmentation of pulmonary embolism in 3D CT pulmonary angiography using a threshold adjustment segmentation network

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 1, 2025

Language: Английский

---

Speed and accuracy in Tandem: Deep Learning-Powered Millisecond-Level pulmonary embolism detection in CTA

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 106, P. 107792 - 107792

Published: March 5, 2025

Language: Английский

---

The Evolution of Computer-assisted Detection of Pulmonary Embolism from Volume to Voxel

Journal of Cardiovascular Emergencies, Journal Year: 2025, Volume and Issue: 11(1), P. 1 - 10

Published: March 1, 2025

Abstract Pulmonary embolism (PE) remains a significant cause of cardiovascular mortality, with untreated cases showing mortality rates up to 30%. The evolution computer-assisted detection (CAD) for PE has transformed dramatically over the past decades, progressing from simple pattern recognition sophisticated deep learning approaches. Early CAD systems demonstrated modest performance, sensitivity around 75% at 2–4 false positives per scan, whereas modern architectures achieve sensitivities 92.9% 0.15 scan. Significantly, technological progression evolved basic patient-level classification voxel-level analysis. This review provides comprehensive overview systems, their clinical value, and future directions.

Language: Английский

---

Advancing Pulmonary Embolism Detection with Integrated Deep Learning Architectures

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown

Published: April 25, 2025

The main aim of this study is to introduce a new hybrid deep learning model for biomedical image classification. We propose novel convolutional neural network (CNN), named HybridNeXt, detecting pulmonary embolism (PE) from computed tomography (CT) images. To evaluate the HybridNeXt model, we created dataset consisting two classes: (1) PE and (2) control. architecture combines different advanced CNN blocks, including MobileNet, ResNet, ConvNeXt, Swin Transformer. specifically designed combine strengths these well-known CNNs. also includes stem, downsampling, output stages. By adjusting parameters, developed lightweight version suitable clinical use. further improve classification performance demonstrate transfer capability, proposed feature engineering (DFE) method using multilevel discrete wavelet transform (MDWT). This DFE has three phases: (i) extraction raw images bands, (ii) selection iterative neighborhood component analysis (INCA), (iii) k-nearest neighbors (kNN) classifier. first trained on training images, creating pretrained model. Then, extracted features applied achieved test accuracy 90.14%, while our improved 96.35%. Overall, results confirm that highly accurate effective presented HybridNeXt-based methods can potentially be other tasks.

Language: Английский

---

Enhancing intima-media complex segmentation with a multi-stage feature fusion-based novel deep learning framework

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108050 - 108050

Published: Feb. 23, 2024

Language: Английский

---

Blood Vessels Disease Detection of Coronary Angiography Images using Deep learning Model

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: June 13, 2024

Presently Coronary artery disease, often caused by the narrowing of coronary lumen due to atherosclerosis, is a leading cause death. angiography also known as cardiac catheterization or X-ray angiography, medical procedure that uses imaging visualize arteries, which supply blood heart muscle. assess flow through these arteries and identify any blockages abnormalities. The accuracy depends on quality equipment well experience expertise radiologist. Poor image could affect accurate diagnosis arteries. Manual interpretation images subjective time consuming. In some cases, small diffuse may not be easily visible, additional techniques required. Therefore, early automated detection blockage vessels became necessary for diagnosis. artificial intelligence algorithms play vital role in this area. paper, deep-learning based algorithm has been used recognition angiographic visuals. Here, we proposed deep learning (YOLOv8) models into images. experiment about 1934 labelled from Mendeley. For Experimentation purpose, are preprocessed augmented. Total 80% have training 20% testing. experimental results show measuring metrices model area rectangular box. performance represented predicted value Precision, recall, mean average precision (mAP) F1 score are, 99.4%, 100%, 99.5% 99.7% respectively.

Language: Английский

---

Enhancing Pulmonary Embolism Segmentation Through Optimized SwinUnet with Resnet 152

Journal of The Institution of Engineers (India) Series B, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 20, 2024

Language: Английский

---