Segmentation and accurate identification of large carious lesions on high quality x-ray images based on Attentional U-Net model. A proof of concept study

Li, Wei; Zhu, Xueyan; Wang, Xiaochun; Wang, Fei; Liu, Junyan; Chen, Mingjun; Wang, Fei; Yue, Honghao

doi:10.1063/5.0084593

Cited by 5 publications

(7 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(to be continued on next page) (533/59/-) PAN 3 dentists (+10y) yes [77] (1250/-/125) PERI dentists (unspecified) yes [81] (3293/252/141) BTW 4 dentists (+3y) yes [82] (492/65/64) BTW 1 dentist + 1 radiologist (+9y) yes [15] (800/-/200) BTW 2 dentists (+10y) yes [83] photos + 120 x-rays * [57] Photos + X-Rays 2 dentists (+3y) yes [1] (1071/-/89) PAN 4 dentists (3-15y) yes [88] BTW 1 experienced dentist in oral radiology yes [38] PAN radiologists (unspecified) yes [34] (30/11/10) PAN 1 dentist + 1 radiologist yes [40] (447/127/61) PAN -yes [37] (485/69/139) PERI 2 dentists + 1 resident dentist yes [61] (80/-/20) PAN 2 dentists yes [84] (400/50/50) BTW 1 radiologist (11y) and a research assistant (3y) yes [62] (1200/150/150) PAN Clinicians (unspecified) yes [85] (935/117/117) PERI 1 radiologist (12y) and a research assistant (2y) yes [51] (76/-/32) PAN 3 dentists (+3y) yes [52] (980/-/420) PAN 3 dentists (+3y) yes [63] (90/-/10) PAN Radiologists (unspecified, +5y) yes [53] (1104/111/121) RVG Dentists (unspecified) yes [89] (175/-/75) PAN Dentists (unspecified, +5y) yes [87] (457/-/195) Unspecified 4 dentists yes [78] (193/83/1224) PAN -no [66] (2507/835/835) PAN 1 dentist + 1 dental student (last year) yes [79] PANs + 682 PERI/BTW * [3] PAN + PERI + BTW 3 dentists yes [86] PAN dentists (unspecified) yes [35] (1005/335/335) BTW 2 dentists yes [80] (1000/0/200) BTW Dentists (unspecified) no [41] ( (iii) number and expertise of professionals involved in the annotation task and (iv) whether the paper presents any information about the dataset annotation protocol. This information is used to investigate RQ1 Concerning RQ1.1, the selected papers exploited three different types of radiograph: panoramic (forty-six papers, 66.67%), periapical (nine papers, 13.04%), and bitewing (eight papers, 11.59%).…”

Section: Datasetsmentioning

confidence: 99%

“…One study (1.45%) combined panoramic and periapical [45], another combined periapical and bitewing [74], another combined panomaric, periapical and bitewing [79], another used RadioVisioGraphy (RVG) [53] and another manipulated panoramic and high-definition photographs of the mouth [83]. One article did not specify the type of radiograph [87]. The average number of samples per dataset is 1234 images.…”

Section: Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning to Detect and Classify Teeth, Dental Caries and Restorations: A Systematic Mapping

Carneiro,

Zancan,

Tirapelli

et al. 2023

Preprint

View full text Add to dashboard Cite

Purpose: Oral health is an indicator of individuals' health and quality of life. Consequently, it is a concerning topic faced everyday by healthcare professionals. X-ray images are essential resources for detecting oral problems. Artificial Intelligence has gained attention in the detection of oral diseases and conditions. We carry out a systematic mapping in the literature regarding the application of deep learning in dental radiographs for detection, segmentation, and classification of teeth, caries, and restorations. Methods: Through automated searches in the ACM Digital Library, IEEE Xplore Digital Library, PubMed, and Scopus, 394 primary articles were found, published between 2012-2023. After applying inclusion and exclusion criteria, 69 articles were read and analyzed considering: consistency and adequacy of the used databases, implemented techniques, and results. Results: 40.82% of the analyzed articles did not present clear information on the approval of ethics committees to carry out the research. Although the application of computational techniques in oral health is an interdisciplinary topic, 37.68% of the 69 surveys were conducted by teams composed only of professionals from one area. 66.67% of the articles explored only panoramic radiographs and the most used metrics were accuracy, recall, and precision. The U-Net and Mask R-CNN networks were the most applied. Conclusion: Despite the increase in the number of investigations, some challenges, such as the low availability of public data sets, the lack of details in the developed methodologies, and the lack of standardization when presenting results, complicate a fair comparison between papers, which is an obstacle to be overcome.

show abstract

Section: Datasetsmentioning

confidence: 99%

Section: Datasetsmentioning

confidence: 99%

Deep Learning to Detect and Classify Teeth, Dental Caries and Restorations: A Systematic Mapping

Carneiro,

Zancan,

Tirapelli

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In this paper, the binary cross entropy of pixels was used as the target to train the U-Net model [16][17].…”

Section: Image Segmentation Using Residual Double Attention U-net Modelmentioning

confidence: 99%

Segmentation for Athlete's Ankle Injury Image Using Residual Double Attention U-Net Model

Zhang,

Zhou,

Huang

et al. 2023

Braz. arch. biol. technol.

View full text Add to dashboard Cite

The image of an athlete's ankle joint injury can help to check whether the athlete's ankle joint is damaged, and plays a very important role in clinical diagnosis. To address the problem of poor segmentation effect of traditional athletes' ankle injury image segmentation algorithm, an ankle injury image segmentation algorithm based on residual double attention U-Net model is proposed. First, the region of interest is extracted from the original ankle injury image. After translation, rotation and turnover, the image data is expanded. Second, the residual structure is used to adjust the gradient propagation and residual feedback of the segmentation framework, extract the attribute information in the region of interest, and combine the two to retain more image features. Finally, combined with the double attention module to improve the weight ratio of image features, the athlete ankle injury image segmentation is realized in the image segmentation framework based on residual double attention U-Net model. The results demonstrate that the maximum values of DSC, ASSD, PM, and CR for the proposed algorithm are 0.93, 0.1, 0.96, and 0.95, respectively, and the F1 score is 95.7%, indicating that the segmentation effect of this algorithm is closer to the theoretical segmentation effect, and higher precision in segmentation, and the segmented image has a high degree of similarity to the original image, resulting in excellent segmentation performance. HIGHLIGHTS• We propose a segmentation using the Residual Double Attention U-Net model.• Adjusting the gradient propagation of the segmentation framework using the residual structure.• Solved the problem of low Correspondence Ratio and F1 values in traditional algorithms.• Using multiple data sets to test the application effect of proposed algorithm.

show abstract

“…Self-attention was one of the main reasons for the emergence of transformers in the Natural Language (NLP) domain which had a huge impact on benchmarking language model due to it being a cutting-edge technology that is now state-of-the-art in the NLP domain. The transformer model was primarily developed for neural machine translation by A. Vaswani et al [25] and then later modified by A. Dosovitskiy et al [29] to work on visual data making the idea of multi-headed self-attention possible for images without the use of convolutions. Since then, there have been numerous variations of transformer model used for computer vision applications like image classification, object detection, and image segmentation.…”

Section: Attention-based and Transformer Basedmentioning

confidence: 99%

Automatic Teeth Segmentation Using Attention U-Net

Mahran¹,

Hussein²,

Saber³

2023

Preprint

View full text Add to dashboard Cite

Teeth segmentation plays a pivotal role in dental diagnosis, treatment, planning, and the development of computer-aided dental systems. It enables precise identification and analysis of dental structures, aiding in detecting dental abnormalities, measuring tooth dimensions, and assessing oral health conditions. Accurate teeth segmentation also facilitates the automation of dental workflows, leading to improved efficiency and reduced human error. Artificial Intelligence (AI) has witnessed rapid advancements, with various approaches developed or enhanced. While Convolutional Neural Networks (CNNs) have been widely used in medical image segmentation, the U-Net architecture has emerged as a standout performer due to its exceptional segmentation capabilities. This paper presents a proof of concept for the Attention U-Net architecture, as previously used in [1], applied to teeth segmentation. The study demonstrates the superior performance of this network in accurately segmenting teeth using a newly available benchmark dataset called Tufts Dental X-Ray Dataset. When trained and tested on 10-fold cross-validation, the model achieved an average dice coefficient of 95.01%, intersection over union of 90.6%, and pixel accuracy of 98.82%. These scores surpass those of all other networks implemented on the same dataset. By leveraging the Attention U-Net architecture, our research showcases the potential of advanced AI techniques in dental radiography. The findings contribute to the ongoing efforts to develop automated systems to assist dental professionals in their clinical practice.

show abstract

Segmentation and accurate identification of large carious lesions on high quality x-ray images based on Attentional U-Net model. A proof of concept study

Cited by 5 publications

References 26 publications

Deep Learning to Detect and Classify Teeth, Dental Caries and Restorations: A Systematic Mapping

Deep Learning to Detect and Classify Teeth, Dental Caries and Restorations: A Systematic Mapping

Segmentation for Athlete's Ankle Injury Image Using Residual Double Attention U-Net Model

Automatic Teeth Segmentation Using Attention U-Net

Contact Info

Product

Resources

About