Explainable Transfer Learning-Based Deep Learning Model for Pelvis Fracture Detection

Kassem, Mohamed A.; Naguib, Soaad M.; Hamza, Hanaa M.; Fouda, Mostafa M.; Saleh, M.; Hosny, Khalid M.

doi:10.1155/2023/3281998

Cited by 31 publications

(2 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When combined with the class activation map, CNNs have a substantial locating capability. CNNs have a significant locating capacity when used with the class activation map [ [21] , [22] , [23] , [24] , [25] ]. A class activation map was used to distinguish the discriminative ROI.…”

Section: Methodsmentioning

confidence: 99%

Automated system for classifying uni-bicompartmental knee osteoarthritis by using redefined residual learning with convolutional neural network

Naguib,

Kassem,

Hamza

et al. 2024

Heliyon

Self Cite

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Automated system for classifying uni-bicompartmental knee osteoarthritis by using redefined residual learning with convolutional neural network

Naguib,

Kassem,

Hamza

et al. 2024

Heliyon

Self Cite

View full text Add to dashboard Cite

“…Mohamed et al [12] developed a unique deep-learning model based on ResNet50 for the easy identification of pelvis fractures in scanning images. The study investigates the pelvis classification problem using convolutional layers and receptor fields, with a GPU being used to increase performance.…”

Section: Related Workmentioning

confidence: 99%

Detection of Hand Bone Fractures in X-Ray Images Using Hybrid YOLO NAS

Medaramatla,

Samhitha,

Pande

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The majority of bones that have fractured in humans are hand bones. As we use our hands widely, they need early and accurate detection to be diagnosed. Fractures in the hands are most frequently brought on by blunt force trauma, sports injuries, and bone fragility. Getting an X-ray of the affected area of the bone and then discussing the results with a medical practitioner or radiologist is the standard procedure for determining whether or not a fracture exists in the bone. The majority of medical professionals and radiologists use X-rays to diagnose hand fractures; however, in some instances, they might miss small or hairline fractures. Additionally, it might be difficult to find a good radiologist who can detect the fracture properly and in time, because a delay in diagnosis can cause the injury to be more severe, and the bone might not be recovered properly. Therefore, to detect hand bone and joint fractures through X-rays, a hybrid model was developed that uses deep learning algorithms YOLO NAS, Efficient Det, and DETR3, which are widely recognized for their exact object detection capabilities. The dataset used for this model is a hybrid dataset of 4736 hand-bone X-ray images, they were further classified into 6 classes based on their types. To evaluate the performance the best method is to compare the proposed model with the existing models, hence, the model was compared with various existing algorithms and result analysis was done.

show abstract

Explainable multi-layer COSFIRE filters robust to corruptions and boundary attack with application to retina and palmprint biometrics

Apap,

Bhole,

Fernández-Robles

et al. 2024

Neural Comput & Applic

View full text Add to dashboard Cite

We propose a novel and versatile computational approach, based on hierarchical COSFIRE filters, that addresses the challenge of explainable retina and palmprint recognition for automatic person identification. Unlike traditional systems that treat these biometrics separately, our method offers a unified solution, leveraging COSFIRE filters’ trainable nature for enhanced selectivity and robustness, while exhibiting explainability and resilience to decision-based black-box adversarial attack and partial matching. COSFIRE filters are trainable, in that their selectivity can be determined with a one-shot learning step. In practice, we configure a COSFIRE filter that is selective for the mutual spatial arrangement of a set of automatically selected keypoints of each retina or palmprint reference image. A query image is then processed by all COSFIRE filters and it is classified with the reference image that was used to configure the COSFIRE filter that gives the strongest similarity score. Our approach, tested on the VARIA and RIDB retina datasets and the IITD palmprint dataset, achieved state-of-the-art results, including perfect classification for retina datasets and a 97.54% accuracy for the palmprint dataset. It proved robust in partial matching tests, achieving over 94% accuracy with 80% image visibility and over 97% with 90% visibility, demonstrating effectiveness with incomplete biometric data. Furthermore, while effectively resisting a decision-based black-box adversarial attack and impervious to imperceptible adversarial images, it is only susceptible to highly perceptible adversarial images with severe noise, which pose minimal concern as they can be easily detected through histogram analysis in preprocessing. In principle, the proposed learning-free hierarchical COSFIRE filters are applicable to any application that requires the identification of certain spatial arrangements of moderately complex features, such as bifurcations and crossovers. Moreover, the selectivity of COSFIRE filters is highly intuitive; and therefore, they provide an explainable solution.

show abstract

Explainable Transfer Learning-Based Deep Learning Model for Pelvis Fracture Detection

Cited by 31 publications

References 32 publications

Automated system for classifying uni-bicompartmental knee osteoarthritis by using redefined residual learning with convolutional neural network

Automated system for classifying uni-bicompartmental knee osteoarthritis by using redefined residual learning with convolutional neural network

Detection of Hand Bone Fractures in X-Ray Images Using Hybrid YOLO NAS

Explainable multi-layer COSFIRE filters robust to corruptions and boundary attack with application to retina and palmprint biometrics

Contact Info

Product

Resources

About