ResMIBCU-Net: an encoder–decoder network with residual blocks, modified inverted residual block, and bi-directional ConvLSTM for impacted tooth segmentation in panoramic X-ray images

Imak, Andaç; Çelebi, Adalet; Polat, Onur; Türkoğlu, Muammer; Şengür, Abdulkadir

doi:10.1007/s11282-023-00677-8

Cited by 11 publications

(6 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They compared a two-stage technique (Faster RCNN with ResNet50, AlexNet, and VGG16 as backbones) with a one-stage technique (YOLOv3) and found that YOLOv3 had the highest detection efficacy with an average precision of 0.96. Imak et al 38 used ResMIBCU-Net to segment impacted teeth (including impacted canines) on panoramic images and achieved an accuracy of 99.82%. Orhan et al 39 evaluated the diagnostic performance of a U-Net CNN model for detecting impacted third molar teeth on CBCT images and showed an accuracy of 86.2%.…”

Section: Discussionmentioning

confidence: 99%

Deep learning driven segmentation of maxillary impacted canine on cone beam computed tomography images

Swaity,

Elgarba,

Morgan

et al. 2024

Sci Rep

View full text Add to dashboard Cite

The process of creating virtual models of dentomaxillofacial structures through three-dimensional segmentation is a crucial component of most digital dental workflows. This process is typically performed using manual or semi-automated approaches, which can be time-consuming and subject to observer bias. The aim of this study was to train and assess the performance of a convolutional neural network (CNN)-based online cloud platform for automated segmentation of maxillary impacted canine on CBCT image. A total of 100 CBCT images with maxillary canine impactions were randomly allocated into two groups: a training set (n = 50) and a testing set (n = 50). The training set was used to train the CNN model and the testing set was employed to evaluate the model performance. Both tasks were performed on an online cloud-based platform, ‘Virtual patient creator’ (Relu, Leuven, Belgium). The performance was assessed using voxel- and surface-based comparison between automated and semi-automated ground truth segmentations. In addition, the time required for segmentation was also calculated. The automated tool showed high performance for segmenting impacted canines with a dice similarity coefficient of 0.99 ± 0.02. Moreover, it was 24 times faster than semi-automated approach. The proposed CNN model achieved fast, consistent, and precise segmentation of maxillary impacted canines.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep learning driven segmentation of maxillary impacted canine on cone beam computed tomography images

Swaity,

Elgarba,

Morgan

et al. 2024

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Imak et al 11 proposed an improved tooth segmentation method based on U-Net,which named ResMIBCU-Net. ResMIBCU-Net uses an improved inverse residual block structure to reduce the semantic gap.…”

Section: Improved Methods Based On U-netmentioning

confidence: 99%

“…[7][8][9] Due to the rapid development of deep learning technology, since the U-Net model was proposed in 2015, 10 the network structure based on encoder and decoder has successfully realized the multilevel feature representation and fusion of medical images, which has led to a great improvement in the segmentation accuracy of medical images. Some works [11][12][13][14] made improvements on the basis of U-Net to achieve accurate dental segmentation. Imak et al 11 achieved 99.8% accuracy on a public dental dataset containing 1500 panoramic dental images.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning-based tooth segmentation methods in medical imaging: A review

Chen,

Ma,

et al. 2024

Proc Inst Mech Eng H

View full text Add to dashboard Cite

Deep learning approaches for tooth segmentation employ convolutional neural networks (CNNs) or Transformers to derive tooth feature maps from extensive training datasets. Tooth segmentation serves as a critical prerequisite for clinical dental analysis and surgical procedures, enabling dentists to comprehensively assess oral conditions and subsequently diagnose pathologies. Over the past decade, deep learning has experienced significant advancements, with researchers introducing efficient models such as U-Net, Mask R-CNN, and Segmentation Transformer (SETR). Building upon these frameworks, scholars have proposed numerous enhancement and optimization modules to attain superior tooth segmentation performance. This paper discusses the deep learning methods of tooth segmentation on dental panoramic radiographs (DPRs), cone-beam computed tomography (CBCT) images, intro oral scan (IOS) models, and others. Finally, we outline performance-enhancing techniques and suggest potential avenues for ongoing research. Numerous challenges remain, including data annotation and model generalization limitations. This paper offers insights for future tooth segmentation studies, potentially facilitating broader clinical adoption.

show abstract

“…Imak ve ark., derin evrişimli sinir ağları kullanarak panoramik diş görüntülerinde gömülü dişlerin tespiti için U-net modeli geliştirmişlerdir. Önerdikleri yöntem %99,82 doğruluk, %91,59 F1 Skor, %84,48 IoU Skor ve %90,71 oranlarında Duyarlılık değerlerini elde etmişlerdir [13].…”

unclassified

ResNet Tabanlı PSPNet Kullanarak Panoramik Görüntülerde Gömülü Diş Segmentasyon Analizi

Durmuş,

Ergen,

Çelebi

et al. 2024

Fırat Üniversitesi Mühendislik Bilimleri Dergisi

Self Cite

View full text Add to dashboard Cite

Diş sağlığı, genel sağlık ve yaşam kalitesi üzerinde önemli bir etkiye sahiptir. Gömülü dişlerin segmentasyonu, diş hekimliğinde erken teşhis ve tedavi için kritik öneme sahip bir adımdır. Bu çalışmada, panoramik diş görüntülerindeki gömülü dişlerin doğru bir şekilde tanımlanması amacıyla derin öğrenme tekniklerinin kullanılması ele alınmıştır. Bu kapsamda, gömülü diş segmentasyonu için ResNet omurga ağına dayalı Piramit Sahne Ayrıştırma Ağı (PSPNet) geliştirilmiştir. Önerilen mimaride, önceden eğitilmiş ResNet omurga ağının ResNet18, ResNet34, ResNet50, ResNet101 ve ResNet152 versiyonları adapte edilmiştir. Bu çalışmada elde edilen bulgular göz önüne alındığında, diş görüntülerindeki segmentasyon ve tanıma süreçlerinde en yüksek başarıyı ResNet18 modeli ile elde edilmiştir (%92.09 F1 Skor, %93.88 Kesinlik, %90.39 Duyarlılık, %85.34 IoU Skor ve %96.89 Dice Katsayısı). Bu araştırma, panoramik diş görüntüleri üzerinde yapılan çalışmalar sonucunda, yetişkin hastalarda gömülü dişlerin başarıyla tespit edilme oranının yüksek olduğunu ortaya koymaktadır. Bu bulgular, yapay zekanın diş hekimleri için etkili bir yardımcı araç olabileceğini vurgulamakta ve sağlık sektöründeki yapay zeka gelişimine olan güveni artırmaktadır.

show abstract

ResMIBCU-Net: an encoder–decoder network with residual blocks, modified inverted residual block, and bi-directional ConvLSTM for impacted tooth segmentation in panoramic X-ray images

Cited by 11 publications

References 34 publications

Deep learning driven segmentation of maxillary impacted canine on cone beam computed tomography images

Deep learning driven segmentation of maxillary impacted canine on cone beam computed tomography images

Deep learning-based tooth segmentation methods in medical imaging: A review

ResNet Tabanlı PSPNet Kullanarak Panoramik Görüntülerde Gömülü Diş Segmentasyon Analizi

Contact Info

Product

Resources

About