Breakthrough in Dentistry: AI model identifies root canal morphology in fused-rooted mandibular second molars in new research
China: In a significant stride for dental diagnostics, researchers have unveiled a pioneering deep learning algorithm capable of identifying root canal morphology in fused-rooted mandibular second molars directly from X-ray images. This breakthrough holds immense promise for improving the accuracy and efficiency of root canal treatments, a cornerstone of modern dentistry.
A recent study published in the Journal of Endodontics found convolutional neural networks (CNNs) more effective than endodontic residents in identifying the three-dimensional root canal morphology of mandibular second molars (MSMs). The result indicates that CNNs can employ two-dimensional images effectively in aiding three-dimensional diagnoses.
Root canal treatment, also known as endodontic therapy, involves meticulously cleaning and shaping the intricate network of root canals within a tooth to alleviate pain and prevent infection. However, the fused root presence complicates this process, often leading to diagnostic errors and suboptimal treatment outcomes.
Understanding the intricate anatomical morphology of fused-rooted MSMs is important for root canal treatment. Weiwei Wu, Huazhong University of Science and Technology, Wuhan, China, and colleagues utilized a deep learning approach to identify the three-dimensional root canal morphology of MSMs from two-dimensional X-ray images.
For this purpose, the researchers included 271 fused-rooted MSMs. Micro-computed tomography (micro-CT) reconstruction and two-dimensional X-ray projection images were obtained. The ground truth of three-dimensional root canal morphology was determined through micro-CT images, classified into symmetrical, merging, and asymmetrical types. The researchers employed traditional augmentation techniques from the Python package Augmentor and a multi-angle projection method to amplify the X-ray image dataset.
The team conducted root canal morphology identification. The classification results from convolutional neural networks (CNNs) were compared with endodontic residents.
The following were the key findings of the study:
· The multi-angle projection augmentation method outperformed the traditional approach in all CNNs except for EfficientNet-b5.
· ResNet18 combined with the multi-angle projection method outperformed all other combinations, with an overall accuracy of 79.25%.
· In specific classifications, 81.13%, 86.79%, and 90.57% accuracies were achieved for merging, symmetrical, and asymmetrical types, respectively.
· CNNs surpassed endodontic residents in classification performance; the average accuracy for endodontic residents was only 60.38%.
In conclusion, utilizing advanced deep learning techniques, the study delved into the root canal morphology of fused-rooted mandibular second molars using two-dimensional X-ray images.
"The model based on convolutional neural networks (CNNs) excelled in identifying symmetrical, merging, and asymmetrical root canal types, outperforming endodontic residents by a significant margin," the researchers wrote.
Researchers are optimistic about further refining the algorithm's performance and exploring its application to other facets of dental diagnostics and treatment planning. With continued collaboration between dental professionals and AI experts, the future of dentistry promises to be brighter than ever, offering enhanced solutions to address the complexities of oral health with unprecedented accuracy and efficiency.
Reference:
Wu W, Chen S, Chen P, Chen M, Yang Y, Gao Y, Hu J, Ma J. Identification of root canal morphology in fused-rooted mandibular second molars from X-ray images based on deep learning. J Endod. 2024 May 29:S0099-2399(24)00338-8. doi: 10.1016/j.joen.2024.05.014. Epub ahead of print. PMID: 38821263.
Disclaimer: This website is primarily for healthcare professionals. The content here does not replace medical advice and should not be used as medical, diagnostic, endorsement, treatment, or prescription advice. Medical science evolves rapidly, and we strive to keep our information current. If you find any discrepancies, please contact us at corrections@medicaldialogues.in. Read our Correction Policy here. Nothing here should be used as a substitute for medical advice, diagnosis, or treatment. We do not endorse any healthcare advice that contradicts a physician's guidance. Use of this site is subject to our Terms of Use, Privacy Policy, and Advertisement Policy. For more details, read our Full Disclaimer here.
NOTE: Join us in combating medical misinformation. If you encounter a questionable health, medical, or medical education claim, email us at factcheck@medicaldialogues.in for evaluation.