<scp>A</scp> machine learning‐based method for automatic diagnosis of ankle fracture using X‐ray images

Wu, Mengkun; Sun, Hao; Sun, Zuoliang; Guo, Xin; Duan, Lunhui; Tan, Yinglun; Cui, Rui

doi:10.1002/ima.22665

Cited by 5 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we introduced a dataset of annotated ankle x-rays and demonstrated the effect of its multiclass labeling by performing network training on the task of automated fracture detection on distinct subsets of images, thus quantifying the influence of selected image features. By utilizing customized state-of-the-art preprocessing and augmentation methods on both the images themselves as well as the composition of the training data set, our study performed better compared to most contemporary ankle studies, including both convolutional neural networks [ 16 , 17 ] and traditional machine learning methods [ 18 ]. For convolutional neural networks, the prevalent training protocol incorporates random dataset augmentation as performed in our study as well as triplication of the medical image for use on imagenet-pretrained networks.…”

Section: Discussionmentioning

confidence: 99%

Multiclass datasets expand neural network utility: an example on ankle radiographs

et al. 2023

View full text Add to dashboard Cite

Purpose Artificial intelligence in computer vision has been increasingly adapted in clinical application since the implementation of neural networks, potentially providing incremental information beyond the mere detection of pathology. As its algorithmic approach propagates input variation, neural networks could be used to identify and evaluate relevant image features. In this study, we introduce a basic dataset structure and demonstrate a pertaining use case. Methods A multidimensional classification of ankle x-rays (n = 1493) rating a variety of features including fracture certainty was used to confirm its usability for separating input variations. We trained a customized neural network on the task of fracture detection using a state-of-the-art preprocessing and training protocol. By grouping the radiographs into subsets according to their image features, the influence of selected features on model performance was evaluated via selective training. Results The models trained on our dataset outperformed most comparable models of current literature with an ROC AUC of 0.943. Excluding ankle x-rays with signs of surgery improved fracture classification performance (AUC 0.955), while limiting the training set to only healthy ankles with and without fracture had no consistent effect. Conclusion Using multiclass datasets and comparing model performance, we were able to demonstrate signs of surgery as a confounding factor, which, following elimination, improved our model. Also eliminating pathologies other than fracture in contrast had no effect on model performance, suggesting a beneficial influence of feature variability for robust model training. Thus, multiclass datasets allow for evaluation of distinct image features, deepening our understanding of pathology imaging.

show abstract

Section: Discussionmentioning

confidence: 99%

Multiclass datasets expand neural network utility: an example on ankle radiographs

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Imaging serves as the principal method for diagnosing orthopedic conditions, including fractures, osteoarthritis, bone tumors, etc. [9]. Misdiagnosis, often due to image misinterpretation or misjudgment, is prevalent in clinical settings.…”

Section: Introductionmentioning

confidence: 99%

YOLOv8-Seg: A Deep Learning Approach for Accurate Classification of Osteoporotic Vertebral Fractures

Yang,

Qian,

Xiao

et al. 2024

Preprint

View full text Add to dashboard Cite

The abstract of the article presents a study focused on the application of deep learning for the classification of osteoporotic vertebral fractures (OVF), a growing health concern among the elderly. The research aimed to explore the potential of deep learning to assist in diagnosing OVF, evaluate the clinical viability of this method, and enhance recovery rates. A dataset comprising 643 CT images of OVF from patients admitted between March 2013 and May 2023 was collected and classified according to the European Vertebral Osteoporosis Study Group (EVOSG) spine classification system. Of these, 613 images were utilized for training and validating a deep learning model, while 30 images served as a test set to assess the model's performance against clinician diagnoses. The deep learning system achieved an impressive 85.9% accuracy rate in classifying fractures according to the EVOSG criteria. The study concludes that deep learning offers a high degree of accuracy in identifying OVF from CT images, which could streamline and improve the current manual diagnostic process that is often complex and challenging. The study also introduces the YOLOv8-Seg model, a novel classification method designed to enhance the diagnostic capabilities for OVF. The use of deep learning in this context is positioned as a significant advancement with the potential to support medical professionals in making early and precise diagnoses, thereby improving patient outcomes. Key terms highlighted in the abstract include deep learning, osteoporotic vertebral fracture, and YOLOv8, indicating the integration of advanced technology in medical diagnosis.

show abstract

Lightweight Deep Learning Model Optimization for Medical Image Analysis

Al‐Milaji,

Yousif

2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Medical image labeling requires specialized knowledge; hence, the solution to the challenge of medical image classification lies in efficiently utilizing the few labeled samples to create a high‐performance model. Building a high‐performance model requires a complicated convolutional neural network (CNN) model with numerous parameters to be trained which makes the test quite expensive. In this paper, we propose optimizing a lightweight deep learning model with only five convolutional layers using the particle swarm optimization (PSO) algorithm to find the best number of kernel filters for each convolutional layer. For colored red, green, and blue (RGB) images acquired from different data sources, we suggest using stain separation using color deconvolution and horizontal and vertical flipping to produce new versions that can concentrate the representation of the images on structures and patterns. To mitigate the effect of training with incorrectly or uncertainly labeled images, grades of disease could have small variances, we apply a second‐pass training excluding uncertain data. With a small number of parameters and higher accuracy, the proposed lightweight deep learning model optimization (LDLMO) algorithm shows strong resilience and generalization ability compared with most recent research on four MedMNIST datasets (RetinaMNIST, BreastMNIST, DermMNIST, and OCTMNIST), Medical‐MNIST, and brain tumor MRI datasets.

show abstract

A machine learning‐based method for automatic diagnosis of ankle fracture using X‐ray images

Cited by 5 publications

References 23 publications

Multiclass datasets expand neural network utility: an example on ankle radiographs

Multiclass datasets expand neural network utility: an example on ankle radiographs

YOLOv8-Seg: A Deep Learning Approach for Accurate Classification of Osteoporotic Vertebral Fractures

Lightweight Deep Learning Model Optimization for Medical Image Analysis

Contact Info

Product

Resources

About