AI-Driven Imaging Techniques In Orthodontics

Was ist der PDCA Zyklus?

Medical imaging is an integral component of the modern healthcare. Advanced imaging techniques allow physicians and surgeons to visualize internal structures without intervention. The latest methods allow you to differentiate between hard and soft tissues. The field of orthodontics focuses on occlusion, skeletal growth, angle of the jaws, and other factors of facial profile. To accurately analyze these features, orthodontists have long been dependent on conventional radiographs like cephalometric X-rays and orthopentograms (OPGs). Luckily, dental imaging has seen the light of revolution and we now see AI being implemented into orthodontic imagery.

Radiomics is a field of research that explores the extraction of radiometric features from images. These metrics are features present within the medical image that help AI identify anatomical points, diagnose underlying disorders, and offer effective treatment plans. With improvements in every iteration, AI is capable of driving imaging techniques in orthodontics. Thus, today, we shall discuss the influence of AI on different types of imaging techniques in orthodontics.

Incorporating AI In Imaging Techniques Of Orthodontics

Cephalometric X-rays

Lateral cephalometric X-rays are the conventional way of diagnosis in orthodontics. This primitive mode of dental imaging isn’t directly influenced by Artificial intelligence. However, AI can improve the shortcomings of this old technique. For example, certain AI programs are capable of improving image quality in medical X-rays. One study showed that AI effectively reduces noise in oral and maxillofacial radiology.

Programs like Carestream (with Smartgrid software) do a tremendous job of identifying noise, reducing it, and providing a better contrast-to-noise ratio in an image. In addition to this, the use of AI in dental-maxillofacial radiology enhances dentists to identify anomalies that are hidden from the naked eye. Reports show that AI in DMFR X-rays can quickly and accurately point out issues like cysts/tumors and periodontal lesions, etc.

Different AI-based programs like WeDoCeph, WebCeph, and CephX are easily accessible through your mobile devices. These multi-functional apps automatically identify landmarks and figure out angles and anatomical distances. Moreover, the software generates a cephalometric report too.

Facial Scans (3D Imaging)

3D imaging is the new standard in orthodontics. Unlike conventional orthodontists who relied mainly on “occlusion” for diagnosis and treatment, the new generation is shifting towards “facial profile analysis”. Now you can get 3D scanners that are efficient and non-invasive tools to get a mouth impression.

The latest equipment are combining 3D facial/intraoral scanning with artificial intelligence and machine learning. According to a study, the first large-scale 3D morphable model (machine learning-based framework) had a diagnostic sensitivity of 95.5% and specificity of 95.2%. The model could automatically detect deformities and provide a treatment plan based on a 3D scan alone.

3D image scanners like the ones provided by Atomica.ai combine intraoral scanning with AI features. The advanced optical sensors capture multiple pictures of the arch while artificial intelligence creates the complete arch picture with high accuracy.

The technology is being widely used in the development of clear aligners. Companies like ClearAligner.ai, etc. utilize data from intraoral scans to plan aligner treatment. The AlignerBot diagnoses and plans the number of orthodontic aligners needed for the perfect smile.

Cone-Bean Computerized Tomography (CBCT)

CBCT scans aid medical diagnosis across the globe. It is the most reliable type of 3-dimensional imaging technique used in healthcare. Cone beam computerized tomography is efficient in orthodontic treatment planning too.

According to studies, advanced AI algorithms can drive CBCT scan leading to better processing of 3D medical images. AI-assisted CBCT scanning plays important roles in three aspects of orthodontic diagnosis and treatment planning:

Detection

AI is quick to identify the anatomical structures and analyze them based on the CNNs. It then identifies the presence of abnormalities such as malocclusions, and jaw growth discrepancies.

Segmentation

The AI programs linked with the CBCT scan determine the exact shape and size of the oral structures (such as teeth). This helps in planning the extent of interproximal reduction or number of extractions required for space creation. Programs like NewTom have a high accuracy in orthodontic diagnosis and treatment planning.

Classification

It also distinguishes the region and extent of the anomaly. Thus, AI algorithms aid health professionals in analyzing CBCT images. Softwares like DIAGNOCAT help dentists reach diagnosis quicker. The advent of AI in CBCT scanning allows the devices to be operated by dental assistants.

Stereophotogrammetry

Stereophotogrammetry is a technique to obtain 3D data for orthodontic purposes. This very technique is capable of reproducing surface geometry with realistic colors and texture data by taking two photographs from different positions. 3D digital stereophotogrammetry is the preferred facial imaging modality for many orthodontists. Artificial intelligence can be paired with photogrammetry to obtain superior results. AI plays a significant role in extracting and scaling data from photogrammetry.

Facial Photography

Intraoral photographs are an essential part of orthodontic treatment. Most orthodontists spend time classifying the pictures for diagnosis and treatment planning. AI-driven orthodontic images can make lives easier. CNN’s (Convolutional neural network) deep learning techniques have shown promising results in segregating photos. A high validity in prediction rate (98%) was seen with AI guidance. The highest prediction rate was seen for:

  • Facial lateral profiles
  • Upper intraoral photos
  • Lower intraoral photos

Final Word

Dental imaging is obligatory for accurate diagnosis of conditions. Orthodontists work with multiple imaging techniques to diagnose and devise treatment plans. In the modern world, Artificial intelligence can shape and drive the commonly used imaging techniques used in orthodontics. AI enhances the picture quality, reduces noise, and identifies important landmarks on 2D cephalometric X-rays. It works best in conjunction with 3D scanning techniques. In modern intraoral scanners, AI can help quickly diagnose the deformity and prescribe the number of clear aligners required for correction. When paired with CBCT, Artificial intelligence accurately detects issues, segments teeth, and classifies deformities. This allows for quicker and better diagnosis. Artificial intelligence also lends a hand in facial photography and advanced stereophotogrammetry.

References

  1. Mayerhoefer, M. E., Materka, A., Langs, G., Häggström, I., Szczypiński, P., Gibbs, P., & Cook, G. (2020). Introduction to radiomics. Journal of Nuclear Medicine, 61(4), 488-495.
  2. Heo, M. S., Kim, J. E., Hwang, J. J., Han, S. S., Kim, J. S., Yi, W. J., & Park, I. W. (2021). Artificial intelligence in oral and maxillofacial radiology: what is currently possible?. Dentomaxillofacial Radiology, 50(3), 20200375.
  3. Hung, K., Montalvao, C., Tanaka, R., Kawai, T., & Bornstein, M. M. (2020). The use and performance of artificial intelligence applications in dental and maxillofacial radiology: A systematic review. Dentomaxillofacial Radiology, 49(1), 20190107.
  4. Knoops, P. G., Papaioannou, A., Borghi, A., Breakey, R. W., Wilson, A. T., Jeelani, O., … & Schievano, S. (2019). A machine learning framework for automated diagnosis and computer-assisted planning in plastic and reconstructive surgery. Scientific reports, 9(1), 13597.
  5. Urban, R., Haluzová, S., Strunga, M., Surovková, J., Lifková, M., Tomášik, J., & Thurzo, A. (2023). AI-assisted CBCT data management in modern dental practice: benefits, limitations and innovations. Electronics, 12(7), 1710.
  6. Tzou, C. H. J., & Frey, M. (2011). Evolution of 3D surface imaging systems in facial plastic surgery. Facial Plastic Surgery Clinics, 19(4), 591-602.
  7. Ryu, J., Lee, Y. S., Mo, S. P., Lim, K., Jung, S. K., & Kim, T. W. (2022). Application of deep learning artificial intelligence technique to the classification of clinical orthodontic photos. BMC Oral Health, 22(1), 454.

You'll find more articles in my blog:

Read more