Maxillofacial fracture detection and classification in computed tomography images using convolutional neural network-based models

Warin, Kritsasith; Limprasert, Wasit; Suebnukarn, Siriwan; Paipongna, Teerawat; Jantana, Patcharapon; Vicharueang, Sothana

doi:10.1038/s41598-023-30640-w

Cited by 28 publications

(7 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…innovatively integrated external attention and 3D feature fusion into YOLOv5 to detect skull fractures in CT images. Warin et al [33] . leveraged YOLOv5 to detect mammofacial fractures in a substantial dataset, classifying fracture conditions into frontal, mid-facial, jaw fractures, and no fractures.…”

Section: Related Workmentioning

confidence: 99%

WCAY Object Detection of Fractures for X-ray Images of Multiple Sites

Chen,

Liu,

et al. 2024

Preprint

View full text Add to dashboard Cite

Presented herein is the WCAY (Weighted Channel attention YOLO) model, meticulously crafted to identify fracture features across diverse X-ray image sites. This model integrates novel core operators and an innovative attention mechanism to enhance its efficacy. Initially, leveraging the benefits of DSConv (Dynamic Snake Convolution), adept at capturing elongated tubular structural features, we introduce the DSC-C2f module to augment the model's fracture detection performance by replacing a portion of C2f. Subsequently, we integrate the newly proposed Weighted Channel attention (WCA) mechanism into the architecture to bolster feature fusion and improve fracture detection across various sites. Comparative experiments were conducted, evaluating the performance of several attention mechanisms. These enhancement strategies were validated through experimentation on public X-ray image datasets (FracAtlas and GRAZPEDWRI-DX). Multiple experimental comparisons substantiate the model's efficacy, demonstrating its superior accuracy and real-time detection capabilities. According to the experimental findings, on the FracAtlas dataset, our WCAY model exhibits a notable 8.8% improvement in mean Average Precision (mAP) over the original model. On the GRAZPEDWRI-DX dataset, the mAP reaches 64.4%, with a detection accuracy of 93.9% for the "fracture" category alone. The proposed model represents a substantial advancement over the original algorithm when compared to other state-of-the-art object detection models. The code is publicly available at https://github.com/cccp421/Fracture-Detection-WCAY .

show abstract

Section: Related Workmentioning

confidence: 99%

WCAY Object Detection of Fractures for X-ray Images of Multiple Sites

Chen,

Liu,

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…In clinical practice, AI has achieved striking success in analyzing patient data like brain-tumor segmentation, 53 assisting in clinical decision-making like epidemiological prediction, 54 and performing complex tasks such as surgery and rehabilitation, which indicates the potential to revolutionize healthcare service. In dentistry, the convolutional neural network has shown performance gain in detecting and classifying maxillofacial fractures from CT. 55 However, subtle details of maxillofacial fractures may not be accurately detected sometimes due to the unfavorable resolution of CT scans. Still, more advanced CT scanners can achieve higher-resolution images in future studies.…”

Section: Ai Technology For Clinical Applicationmentioning

confidence: 99%

ChatGPT for shaping the future of dentistry: the potential of multi-modal large language model

et al. 2023

View full text Add to dashboard Cite

The ChatGPT, a lite and conversational variant of Generative Pretrained Transformer 4 (GPT-4) developed by OpenAI, is one of the milestone Large Language Models (LLMs) with billions of parameters. LLMs have stirred up much interest among researchers and practitioners in their impressive skills in natural language processing tasks, which profoundly impact various fields. This paper mainly discusses the future applications of LLMs in dentistry. We introduce two primary LLM deployment methods in dentistry, including automated dental diagnosis and cross-modal dental diagnosis, and examine their potential applications. Especially, equipped with a cross-modal encoder, a single LLM can manage multi-source data and conduct advanced natural language reasoning to perform complex clinical operations. We also present cases to demonstrate the potential of a fully automatic Multi-Modal LLM AI system for dentistry clinical application. While LLMs offer significant potential benefits, the challenges, such as data privacy, data quality, and model bias, need further study. Overall, LLMs have the potential to revolutionize dental diagnosis and treatment, which indicates a promising avenue for clinical application and research in dentistry.

show abstract

“…Inspired by the artificial model of biological neurons [ 8 ], artificial neural networks have garnered significant attention from researchers as a nonlinear approach for investigating the relationship between input and output information [ 10 ]. Serving as an effective method for general classification, clustering, and prediction, the advantage of utilizing neural networks lies in their ability to adapt to diverse datasets without making any assumptions about the underlying model structure [ 11 , 12 ]. In recent years, neural networks have been applied in various domains related to bone health, including detecting and classifying bone lesions [ 13 ] and assessing bone age [ 11 ].…”

Section: Introductionmentioning

confidence: 99%

“…Serving as an effective method for general classification, clustering, and prediction, the advantage of utilizing neural networks lies in their ability to adapt to diverse datasets without making any assumptions about the underlying model structure [ 11 , 12 ]. In recent years, neural networks have been applied in various domains related to bone health, including detecting and classifying bone lesions [ 13 ] and assessing bone age [ 11 ]. However, there is currently limited utilization of neural network models in exploring factors influencing bone health.…”

Section: Introductionmentioning

confidence: 99%

Opposing Effects of Nutritional Supply on Bone Health at Different Ages: Based on the National Health and Nutrition Examination Survey Database

Wei,

Zhang,

Yuan

et al. 2024

Nutrients

View full text Add to dashboard Cite

(1) Background: Nutrients play an essential role in bone health, whether in achieving peak bone mineral density (BMD) or maintaining bone health. This study explores the relationship between nutrient supply and femoral bone health at different ages. (2) Methods: A total of 5603 participants meeting the inclusion and exclusion criteria were included in this study using the National Health and Nutrition Examination Survey (NHANES) database from 2005 to 2010, 2013 to 2014, and 2017 to 2018. Femoral bone mineral density and bone status were dependent variables, and dietary nutrient intake and nutrient intake status were independent variables. The relationship between dietary nutrient intake and bone mineral density was explored, and the importance of nutrients affecting bone status was analyzed through a neural network model. At the same time, we investigated the relationship between nutrient intake and bone status. (3) Results: The peak of age and femoral bone mineral density appeared at 20 years old in our study. After grouping by age, logistic regression analysis showed that before 20 years old, without adjusting other variables, high-fat diet was more likely to have normal bone mass than appropriate fat diet (OR: 4.173, 95%CI: 1.007–17.289). After adjusting for all demographic factors, niacin intake (OR: 1.062, 95%CI: 1.019–1.108) was beneficial for normal bone mass, while vitamin B6 intake (OR: 0.627, 95%CI: 0.408–0.965) was not. After 20 years old, after adjusting for carbohydrate, protein, vitamin B6, niacin, dietary fat, vitamin B2, and vitamin B12, vitamin B2 intake (OR: 1.153, 95%CI: 1.04–1.278) was beneficial for normal bone mass, while vitamin B6 intake (OR: 0.842, 95%CI: 0.726–0.976) was not. After adjusting for all confounding factors, vitamin B2 intake (OR: 1.288, 95%CI: 1.102–1.506) was beneficial for normal bone mass. In addition, we found that even if there was no statistical significance, the effects of high-fat diet on bone mass were different at different ages. (4) Conclusions: By conducting an in-depth analysis of the NHANES database, this study reveals that dietary factors exert divergent effects on bone health across different age groups, implying the necessity of implementing tailored dietary strategies to maintain optimal bone health at distinct life stages.

show abstract

Maxillofacial fracture detection and classification in computed tomography images using convolutional neural network-based models

Cited by 28 publications

References 29 publications

WCAY Object Detection of Fractures for X-ray Images of Multiple Sites

WCAY Object Detection of Fractures for X-ray Images of Multiple Sites

ChatGPT for shaping the future of dentistry: the potential of multi-modal large language model

Opposing Effects of Nutritional Supply on Bone Health at Different Ages: Based on the National Health and Nutrition Examination Survey Database

Contact Info

Product

Resources

About