Convolutional Neural Network Performance for Sella Turcica Segmentation and Classification Using CBCT Images

Duman, Şuayip Burak; Syed, Ali; Ozen, Duygu Celik; Bayrakdar, İ̇brahim Şevki; Salehi, Hassan S.; Abdel‐Karim, Ahmed; Çelik, Özer; Eser, Gözde; Altun, Oğuzhan; Orhan, Kaan

doi:10.3390/diagnostics12092244

Cited by 15 publications

(8 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CBCT can present anatomical structures in three dimensions, offering clinicians a broader perspective. Subsequently, AI tools can autonomously recognize and segment distinct teeth [13][14][15], skull structures, including the sella turcica [16], maxillary [17], and mandible [18,19]. Additionally, they can identify features such as maxillary sinus mucosa [20], inferior alveolar canal [21], and pharyngeal airway [22,23], within precise three-dimensional (3D) images for diagnostic and printing applications in the medical field.…”

Section: Cone Beam Computed Tomography (Cbct)mentioning

confidence: 99%

Dental Artificial Intelligence Systems: A Review of Various Data Types

Zhang,

Chen,

et al. 2024

Discovery Medicine

View full text Add to dashboard Cite

With the rapid development of dental artificial intelligence systems (DAIS), a new field known as "Data Dentistry", proposed by Schwendicke in 2021, has successfully bridged the gap between medicine and engineering. This literature review introduces advanced techniques in data collection, outlines the current state of DAIS in data processing, and anticipates the future of DAIS by emphasizing the importance of more extensive and enhanced datasets. The key findings include: Versatility of imaging data: Various types of imaging data, such as X-ray, cone beam computed tomography (CBCT), facial photos, and face and oral scans, can be transformed into datasets used by artificial intelligence systems. Uniform rules in electronic dental record (EDR) systems: EDR systems require standardized rules for general use in DAIS, ensuring compatibility and seamless integration. Potential of wearable device data: Data from wearable devices, including bioelectric signals (such as electromyography), stress sensors, AR glasses, etc., show great potential for enhancing DAIS capabilities. Current DAIS performance focus: Presently, DAIS demonstrate superior performance in object location and disease diagnosis compared to information integration and clinical decision-making. Need for data quality and quantity improvement: Further improvements are needed in both the quality and quantity of data for DAIS.

show abstract

Section: Cone Beam Computed Tomography (Cbct)mentioning

confidence: 99%

Dental Artificial Intelligence Systems: A Review of Various Data Types

Zhang,

Chen,

et al. 2024

Discovery Medicine

View full text Add to dashboard Cite

show abstract

“…For frontal sinus segmentation on CBCT images, we used FSNet which had a U-shape encoder-decoder architecture with transfer learning. Five popular backbones, namely VGG16 28 , ResNet101 29 , DenseNet201 30 , Inception V3 31 , and E cientNet-B5 32 were used as encoders in FSNet. The decoder part had ve levels of layers with 2D convolution blocks and a 2D transposed convolution layer for 2D upsampling.…”

Section: The Architecture Of a Two-stage Anatomy-guided Attention Net...mentioning

confidence: 99%

Automatic segmentation and classification of frontal sinuses for sex determination from CBCT scans using a two-stage anatomy-guided attention network: A study of the South Korean population

Silva,

Yang,

Kim

et al. 2023

Preprint

View full text Add to dashboard Cite

Sex determination is essential for identifying unidentified individuals, particularly in forensic contexts. Traditional methods for sex determination involve manual measurements of skeletal features on CBCT scans. However, these manual measurements are labor-intensive, time-consuming, and error-prone. The purpose of this study was to automatically and accurately determine sex on a CBCT scan using a two-stage anatomy-guided attention network (SDetNet). SDetNet consisted of a 2D frontal sinus segmentation network (FSNet) and a 3D anatomy-guided attention network (SDNet). FSNet segmented frontal sinus regions in the CBCT images and extracted regions of interest (ROIs) near them. Then, the ROIs were fed into SDNet to predict sex accurately. To improve sex determination performance, we proposed multi-channel inputs (MSIs) and an anatomy-guided attention module (AGAM), which encouraged SDetNet to learn differences in the anatomical context of the frontal sinus between males and females. SDetNet showed superior sex determination performance in the area under the receiver operating characteristic curve, accuracy, Brier score, and specificity compared with the other 3D CNNs. Moreover, the results of ablation studies showed a notable improvement in sex determination with the embedding of both MSI and AGAM. Consequently, SDetNet demonstrated automatic and accurate sex determination by learning the anatomical context information of the frontal sinus on CBCT scans.

show abstract

“…In a recent CNN-based AI study, Sella Turcica segmentation and classification using CBCT images yielded results with a high percentage of accuracy [64]. Using artificial intelligence algorithms, it can be predicted that the detection of anatomical signs with orthodontic importance will save orthodontists time and facilitate diagnosis.…”

Section: Orthodonticsmentioning

confidence: 99%

Artificial Intelligence in Dentistry

Gokdeniz¹,

Büyüksungur²,

Kolsuz³

2023

Dentistry

View full text Add to dashboard Cite

Artificial intelligence (AI) is the branch of computer science dedicated to building systems to perform tasks that normally require human intelligence. AI tries to solve problems and learn similar to humans. The field of AI has experienced phenomenal development and growth over the past two decades; with the latest developments in digitized data collection, machine learning, and computing infrastructure, AI applications are expanding rapidly, especially in areas that are thought to be reserved for experts in their fields. Artificial intelligence has started to take place rapidly in dental clinical applications. The use of artificial intelligence in dentistry has enormous potential to improve patient care and drive major advances in healthcare. AI in dentistry is being researched for various purposes, such as identifying anatomical and pathological structures, diagnosing diseases and predicting treatment results, and selecting materials to be used. Advances in AI offer healthcare benefits, such as reducing postoperative complications, improving quality of life, and reducing the number of unnecessary procedures. It can also play a great helping role for dentists in increasing the accuracy of diagnosis. This chapter aims to explain the current applications and future predictions of artificial intelligence in dentistry, which is one of the most current topics of recent times.

show abstract

Convolutional Neural Network Performance for Sella Turcica Segmentation and Classification Using CBCT Images

Cited by 15 publications

References 37 publications

Dental Artificial Intelligence Systems: A Review of Various Data Types

Dental Artificial Intelligence Systems: A Review of Various Data Types

Automatic segmentation and classification of frontal sinuses for sex determination from CBCT scans using a two-stage anatomy-guided attention network: A study of the South Korean population

Artificial Intelligence in Dentistry

Contact Info

Product

Resources

About