Imagen de Portada

MSFE-GallNet-X: a multi-scale feature extraction-based CNN Model for gallbladder disease analysis with enhanced explainability

Guardado en:
Publicado en:BMC Medical Imaging vol. 25 (2025), p. 1-22
Autor principal: Hadiur Rahman Nabil
Otros Autores: Istyak Ahmed, Das, Aritra, M. F. Mridha Mohsin Kabir, Aung, Zeyar
Publicado:
Springer Nature B.V.
Materias:
Feature extraction
Accuracy
Gallbladder
Artificial intelligence
Datasets
Deep learning
Performance evaluation
Classification
Image resolution
Disease
Artificial neural networks
Gallbladder cancer
Medical imaging
Gallbladder diseases
Gallstones
Abdomen
Machine learning
Explainable artificial intelligence
Breast cancer
Efficiency
Calculi
Ultrasound
Clinical decision making
Cholecystitis
Parameters
Ultrasonic imaging
Decision making
Neural networks
Acceso en línea:Citation/Abstract
Full Text
Full Text - PDF
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Resumen:Section ObjectiveThis study introduces MSFE-GallNet-X, a domain-adaptive deep learning model utilizing multi-scale feature extraction (MSFE) to improve the classification accuracy of gallbladder diseases from grayscale ultrasound images, while integrating explainable artificial intelligence (XAI) methods to enhance clinical interpretability.AbstractSection MethodsWe developed a convolutional neural network-based architecture that automatically learns multi-scale features from a dataset comprising 10,692 high-resolution ultrasound images from 1,782 patients, covering nine gallbladder disease classes, including gallstones, cholecystitis, and carcinoma. The model incorporated Gradient-Weighted Class Activation Mapping (Grad-CAM) and Local Interpretable Model-Agnostic Explanations (LIME) to provide visual interpretability of diagnostic predictions. Model performance was evaluated using standard metrics, including accuracy and F1 score.AbstractSection ResultsThe MSFE-GallNet-X achieved a classification accuracy of 99.63% and an F1 score of 99.50%, outperforming state-of-the-art models including VGG-19 (98.89%) and DenseNet121 (91.81%), while maintaining greater parameter efficiency, only 1·91 M parameters in gallbladder disease classification. Visualization through Grad-CAM and LIME highlighted critical image regions influencing model predictions, supporting explainability for clinical use.AbstractSection ConclusionMSFE-GallNet-X demonstrates strong performance on a controlled and balanced dataset, suggesting its potential as an AI-assisted tool for clinical decision-making in gallbladder disease management.AbstractSection Clinical trial numberNot applicable.
ISSN:1471-2342
DOI:10.1186/s12880-025-01902-y
Fuente:Health & Medical Collection