Boosting research on dental panoramic radiographs: a challenging data set, baselines, and a task central online platform for benchmark

Silva, Bernardo Peters Menezes; Pinheiro, Laís; Sobrinho, Brenda Pereira Pinheiro; Lima, Fernanda Pereira; Sobrinho, Bruna Pereira Pinheiro; Lima, Kalyf Abdalla Buzar; Pithon, Matheus Melo; Cury, Patricia Ramos; Oliveira, Luciano

doi:10.1080/21681163.2022.2157747

Cited by 5 publications

(6 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…OdontoAI : The OdontoAI platform provides a public dataset with 4,000 PRs, of which 2,000 are annotated [ 19 ]. The annotations include tooth segmentations with corresponding FDI numbers.…”

Section: Methodsmentioning

confidence: 99%

“…Therefore, the current study combined two public datasets (OdontoAI [ 19 ], DENTEX [ 21 ]) to develop an automated method for the novel task of concurrent tooth segmentation, FDI labeling, and diagnosis classification, including caries, impacted tooth, and periapical lesions. This study aimed to evaluate the advantages and challenges of using publicly available datasets in dental AI research, focusing on improving the diagnostic accuracy of caries, impacted teeth, and periapical lesions in PRs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combining public datasets for automated tooth assessment in panoramic radiographs

van Nistelrooij,

Ghoul,

et al. 2024

BMC Oral Health

View full text Add to dashboard Cite

Objective Panoramic radiographs (PRs) provide a comprehensive view of the oral and maxillofacial region and are used routinely to assess dental and osseous pathologies. Artificial intelligence (AI) can be used to improve the diagnostic accuracy of PRs compared to bitewings and periapical radiographs. This study aimed to evaluate the advantages and challenges of using publicly available datasets in dental AI research, focusing on solving the novel task of predicting tooth segmentations, FDI numbers, and tooth diagnoses, simultaneously. Materials and methods Datasets from the OdontoAI platform (tooth instance segmentations) and the DENTEX challenge (tooth bounding boxes with associated diagnoses) were combined to develop a two-stage AI model. The first stage implemented tooth instance segmentation with FDI numbering and extracted regions of interest around each tooth segmentation, whereafter the second stage implemented multi-label classification to detect dental caries, impacted teeth, and periapical lesions in PRs. The performance of the automated tooth segmentation algorithm was evaluated using a free-response receiver-operating-characteristics (FROC) curve and mean average precision (mAP) metrics. The diagnostic accuracy of detection and classification of dental pathology was evaluated with ROC curves and F1 and AUC metrics. Results The two-stage AI model achieved high accuracy in tooth segmentations with a FROC score of 0.988 and a mAP of 0.848. High accuracy was also achieved in the diagnostic classification of impacted teeth (F1 = 0.901, AUC = 0.996), whereas moderate accuracy was achieved in the diagnostic classification of deep caries (F1 = 0.683, AUC = 0.960), early caries (F1 = 0.662, AUC = 0.881), and periapical lesions (F1 = 0.603, AUC = 0.974). The model’s performance correlated positively with the quality of annotations in the used public datasets. Selected samples from the DENTEX dataset revealed cases of missing (false-negative) and incorrect (false-positive) diagnoses, which negatively influenced the performance of the AI model. Conclusions The use and pooling of public datasets in dental AI research can significantly accelerate the development of new AI models and enable fast exploration of novel tasks. However, standardized quality assurance is essential before using the datasets to ensure reliable outcomes and limit potential biases.

show abstract

“…OdontoAI : The OdontoAI platform provides a public dataset with 4,000 PRs, of which 2,000 are annotated [ 19 ]. The annotations include tooth segmentations with corresponding FDI numbers.…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Combining public datasets for automated tooth assessment in panoramic radiographs

van Nistelrooij,

Ghoul,

et al. 2024

BMC Oral Health

View full text Add to dashboard Cite

show abstract

“…Da bi se nadomestio nedostatak podataka, koristiće se vrednosti težinskih koeficijenata modela neuronske mreže koji je obučavan na MSCOCO skupu za obuku. Pored opisanog eksperimenta koji je podrazumevao optimizaciju parametara modifikovane mreže na inicijalnom skupu podataka od 20 slika, u radu su razmatrane još tri eksperimentalne postavke korišćenjem OdontoAI baze slika predložene u [4]. Tako su analizirane performanse modela doobučavanih na skupovima podataka preuzetim iz OdontoAI baze snimaka, tj.…”

Section: Opis Eksperimentalne Postavkeunclassified

“…Zahvaljujemo se doktoru Bojanu Petroviću, redovnom profesoru Medicinskog fakulteta Univerziteta u Novom Sadu, i Klinici za stomatologiju Vojvodine, koji su kroz konsultacije i pripremu panoramskih radiografskih snimaka podržali istraživanje opisano u radu. Takođe se zahvaljujemo i kolegama sa Federalnog univerziteta u Bahiji, u Brazilu, koji su ljubazno ustupili bazu slika [4] za potrebe ovog istraživanja.…”

Section: Zahvalnicaunclassified

Detekcija i segmentacija zuba na panoramskim radiografskim snimcima korišćenjem mašinskog učenja

Petrović

2023

Zbornik radova FTN

View full text Add to dashboard Cite

Panoramski radiografski snimci koriste se u stomatologiji kao pomoćni alat pri prevenciji i dijagnostici zubnih oboljenja. U radu je predložen jedan postupak za automatizovanu detekciju i segmentaciju pojedinačnih zuba na panoramskim radiografskim snimcima vilice. Metod se zasniva na primeni Mask R-CNN algoritma i metodologiji za anotaciju radiografskih snimaka zuba. Razmatrane su strategije mašinskog učenja i performanse sistema zasnovanih na skupovima za obuku različitih veličina. Poređene su karakteristike dobijenih rešenja i potrebnog vremena obuke nad četiri skupa podataka različitih veličina.

show abstract

“…And the annotation needs to consume a lot of time of dental experts, but also highly dependent on the cognition and experience of dental experts, which is undoubtedly unfriendly to deep learning networks that need a large amount of data for training. It is fortunate that some medical scholars have made efforts for this purpose and provided some public DPR datasets 3,[28][29][30] and 3D dental datasets 4 for related research. However, due to the varying parameters of different imaging devices, the same deep learning model may produce large performance differences in different datasets.…”

Section: Introductionmentioning

confidence: 99%

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

Boosting research on dental panoramic radiographs: a challenging data set, baselines, and a task central online platform for benchmark

Cited by 5 publications

References 41 publications

Combining public datasets for automated tooth assessment in panoramic radiographs

Combining public datasets for automated tooth assessment in panoramic radiographs

Detekcija i segmentacija zuba na panoramskim radiografskim snimcima korišćenjem mašinskog učenja

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

Contact Info

Product

Resources

About