Reconstruction-based approach for chest X-ray image segmentation and enhanced multi-label chest disease classification

Artificial Intelligence in Medicine, Journal Year: 2025, Volume and Issue: 165, P. 103135 - 103135

Published: April 23, 2025

U-Net is a commonly used model for medical image segmentation. However, when applied to chest X-ray images that show pathologies, it often fails include these critical pathological areas in the generated masks. To address this limitation, our study, we tackled challenge of precise segmentation and mask generation by developing novel approach, using CycleGAN, encompasses affected pathologies within region interest, allowing extraction relevant radiomic features linked pathologies. Furthermore, adopted feature selection approach focus analysis on most significant features. The results proposed pipeline are promising, with an average accuracy 92.05% AUC 89.48% multi-label classification effusion infiltration acquired from ChestX-ray14 dataset, XGBoost model. applying methodology 14 diseases dataset resulted 83.12%, outperforming previous studies. This research highlights importance effective accurate diseases. promising underscore its potential broader applications

Language: Английский

---

Fast-YOLO Network Model for X-Ray Image Detection of Pneumonia

Electronics, Journal Year: 2025, Volume and Issue: 14(5), P. 903 - 903

Published: Feb. 25, 2025

Pneumonia is a respiratory infection that affects the lungs. The symptoms of viral and bacterial pneumonia are similar. In order to improve automatic detection efficiency regarding X-ray images pneumonia, this paper, we propose novel method based on Fast-YOLO network model. First, re-annotated open-source dataset MIMIC Chest enhancing model’s adaptability complex scenes by incorporating Mixup, Mosaic, Copy–Paste augmentation methods. Additionally, CutMix Random Erasing were introduced increase data diversity. Next, developed lightweight FASPA Fast Pyramid Attention Mechanism designed mechanism effectively address features in images, such as low contrast an uneven distribution local lesions. improves upon YOLOv11 architecture replacing C3k2 module with attention mechanism, significantly reducing network’s parameter count while maintaining performance. Furthermore, enhances feature extraction capabilities when handling geometric deformations, multi-scale features, dynamic changes. It expands receptive field, thereby balancing computational accuracy. Finally, experimental results demonstrate network, compared traditional convolutional neural methods, can identify regions localize lesions image tasks, achieving significant improvements FPS, precision, recall, mAP @0.5, @0.5:0.95. This confirms strikes balance between excellent generalization capability across different datasets has been validated, showing potential accelerate diagnostic process for clinicians enhance

Language: Английский

---

Enhanced super-resolution generative adversarial network augmented convolution neural network for pneumonia prognosis in India: promising health policy implications

International Journal of Systems Assurance Engineering and Management, Journal Year: 2025, Volume and Issue: unknown

Published: April 6, 2025

Language: Английский

---

Development and validation of an interpretable machine learning model for diagnosing pathologic complete response in breast cancer

Computer Methods and Programs in Biomedicine, Journal Year: 2025, Volume and Issue: unknown, P. 108803 - 108803

Published: April 1, 2025

Language: Английский

---

A Modularly Designed Controllable Generative Frameworkfor Glioma and MRI Editing Via Style Representations Enhancement

Published: Jan. 1, 2024

Language: Английский

---