AI-Assisted Orthodontics: A Comprehensive Review

Warum Verantwortung und Feedback für deinen Praxisalltag so wichtig sind…

Artificial intelligence is rapidly becoming an effective tool in the world of healthcare. Deep learning and specialized algorithms in AI are assisting in diagnosis and treatment planning. Doctors have recently seen a surge of AI systems in the medical and dental fields. Orthodontics is a complex field of dentistry that can benefit from artificial intelligence. This review discusses the aspects of orthodontics in which AI can lend a hand.

AI can easily become a part of an orthodontist’s daily routine. You can take advantage of artificial intelligence systems in the following fields of orthodontics:

  • Diagnosis and analysis
  • Treatment planning
  • Research and development

With the incorporation of AI, you can revolutionize the following facets of orthodontic dentistry:

Diagnosis And Analysis

An accurate diagnosis is essential for devising the perfect orthodontic treatment plan. This complex field requires meticulous work on minute details. It is a laborious and time-consuming job. However, AI can make it easy and save precious time.

Landmark Identification (On Cephalogram)

The first and primary step in orthodontic diagnosis is landmark identification on a cephalometric X-ray. Orthodontists perform geometric evaluations (angles, distance, ratios, etc.) based on the landmarks. Orthodontists across the globe have advocated AI assistance in landmark identification, thanks to the highly efficient neural networks (Convolutional and Artificial).

Artificial intelligence programs like YOLOv3 and ACDS have proven to be reliable and time-saving tools for orthodontists. According to a study, the YOLOv3 algorithm successfully and accurately (metric deviation of 2mm) identified 80 landmarks on 283 images. This makes it a viable option for cephalometric landmarking which might be even better than human analysis.

Growth Prediction

As orthodontists mostly deal with growing children, predicting growth and puberty is pivotal. Growth dynamics vary between individuals. Therefore, it is critical to assess the skeletal age of the patient. Reliable methods of assessing skeletal age include the Cervical-Vertebral Method (CVM) and wrist analysis.

Determination of growth and skeletal maturity from the above-mentioned methods can be tricky for the inexperienced. Thus, for all such practitioners, AI can serve as a ray of hope. Deep learning models are capable of achieving more than 90% accuracy in determining CVM stage classification. Thus, it can facilitate dentists and minimize the chances of errors.

Tooth Segmentation

Nowadays, orthodontists carry out tooth segmentations on 3D orthodontic models for better analysis and understanding. This allows precise comprehension of the shape and size of the tooth as well as the bone, which in turn, plays a part in orthodontic tooth movement. AI can do this job for dentists.

Per an authentic study, a convolutional neural network-based AI system performed automated segmentation (with an IoU score of 91%) on intraoral scanned data. It was concluded that AI software is accurate and efficient in segmentation. Moreover, AI-driven novel tools accurately segmented teeth on panoramic radiographs.

Decision-Making

The next step where you can enjoy the perks of artificial intelligence is decision support for orthodontic maneuvers.

Extractions Or No-Extractions

A lot of specialist’s effort and time goes into determining the best treatment modality for the patient. To extract or not extract is a challenging question even for experienced orthodontists.

Fortunately, Artificial intelligence can unburden you by providing accurate therapeutic solutions. It was observed in an initial study that there was a high coincidence (90%) between AI-based treatment plans and those performed by the doctor.

After careful analyses of sagittal discrepancies, crowding, and protrusion, artificial intelligence supports decision-making for orthodontic extraction. According to a review, AI shows promising accuracy in orthodontic decision-making.

The high accuracy can attributed to the accurate identification of parameters like overjet, overbite, crowding, positioning of the anterior teeth lip closure, etc. Based on the model and radiographic data, AI’s predicted treatment plan (extraction/non-extraction) has an accuracy of more than 93%.

It not only aids in decision-making but also provides you with the ideal extraction pattern. The predicted extraction pattern for the first and second upper/lower premolars by AI also has high accuracy.

Orthognathic Surgery

In severe cases with skeletal deformity, camouflage treatment isn’t sufficient. Therefore, orthodontists have to opt for orthognathic surgery. Consideration of various factors is required for adequate conclusion of surgical intervention. The decision to advise surgery is a tough one for doctors because the decision may vary between professionals. There is no standardization to judge the need for surgery. However, AI algorithms are trained based on radiographic (lateral and cephalometric X-rays) and clinical features.

The success rate of the surgery/non-surgery plan by artificial intelligence model+machine learning was found to be 96%. Automated treatment planning for surgery is good. However, the accuracy for class Ⅱ cases was found to be better than that for class Ⅲ malocclusion.

Studies And Research

The use of AI in research and development can potentially speed up the process while minimizing bias. Numerous studies are significant in paving the way for optimum orthodontic treatment. By lending a hand in TMD (temporomandibular joint dysfunction) and cleft studies, AI is opening new avenues for orthodontic treatment.

Conclusion

AI-assisted orthodontics is the new buzz in town. While CHatGPT4.0 is answering questions to every answer, AI systems are also helping orthodontists in diagnosis, treatment planning, and research. Starting with cephalometric analysis, programs like YOLOv3 accurately identify cephalometric landmarks. Advanced algorithms also do a great job of growth prediction based on CVM and wrist analysis. Automated tooth segmentation allows better comprehension of tooth structure.

AI also solves the question of extraction or non-extraction therapy, and that too with high accuracy. So much so, that it also provides the ideal extraction pattern. Artificial intelligence successfully predicts the need for a surgical intervention (orthognathic surgery). Moreover, it facilitates cleft and TMD studies, which can improve treatment.

References

  1. Bohr, A., & Memarzadeh, K. (2020). The rise of artificial intelligence in healthcare applications. In Artificial Intelligence in healthcare (pp. 25-60). Academic Press.
  2. Subramanian, A. K., Chen, Y., Almalki, A., Sivamurthy, G., & Kafle, D. (2022). Cephalometric analysis in orthodontics using artificial intelligence—A comprehensive review. BioMed Research International, 2022(1), 1880113.
  3. Hwang, H. W., Park, J. H., Moon, J. H., Yu, Y., Kim, H., Her, S. B., … & Lee, S. J. (2020). Automated identification of cephalometric landmarks: Part 2-Might it be better than human?. The Angle Orthodontist, 90(1), 69-76.
  4. Seo, H., Hwang, J., Jeong, T., & Shin, J. (2021). Comparison of deep learning models for cervical vertebral maturation stage classification on lateral cephalometric radiographs. Journal of Clinical Medicine, 10(16), 3591.
  5. Ma, T., Yang, Y., Zhai, J., Yang, J., & Zhang, J. (2022, October). A tooth segmentation method based on multiple geometric feature learning. In Healthcare (Vol. 10, No. 10, p. 2089). MDPI.
  6. Wang, X., Alqahtani, K. A., Van den Bogaert, T., Shujaat, S., Jacobs, R., & Shaheen, E. (2024). Convolutional neural network for automated tooth segmentation on intraoral scans. BMC Oral Health, 24(1), 804.
  7. Leite, A. F., Gerven, A. V., Willems, H., Beznik, T., Lahoud, P., Gaêta-Araujo, H., … & Jacobs, R. (2021). Artificial intelligence-driven novel tool for tooth detection and segmentation on panoramic radiographs. Clinical oral investigations, 25, 2257-2267.
  8. Takada, K., Yagi, M., & Horiguchi, E. (2009). Computational Formulation of Orthodontic Tooth-Extraction Decisions: Part I: to extract or not to extract. The Angle orthodontist, 79(5), 885-891.
  9. Evangelista, K., de Freitas Silva, B. S., Yamamoto-Silva, F. P., Valladares-Neto, J., Silva, M. A. G., Cevidanes, L. H. S., … & Massignan, C. (2022). Accuracy of artificial intelligence for tooth extraction decision-making in orthodontics: a systematic review and meta-analysis. Clinical Oral Investigations, 26(12), 6893-6905.
  10. Del Real, A., Del Real, O., Sardina, S., & Oyonarte, R. (2022). Use of automated artificial intelligence to predict the need for orthodontic extractions. Korean Journal of Orthodontics, 52(2), 102-111.
  11. Li, P., Kong, D., Tang, T., Su, D., Yang, P., Wang, H., … & Liu, Y. (2019). Orthodontic treatment planning based on artificial neural networks. Scientific reports, 9(1), 2037.
  12. Choi, H. I., Jung, S. K., Baek, S. H., Lim, W. H., Ahn, S. J., Yang, I. H., & Kim, T. W. (2019). Artificial intelligent model with neural network machine learning for the diagnosis of orthognathic surgery. Journal of Craniofacial Surgery, 30(7), 1986-1989.

You'll find more articles in my blog:

Read more