Search

EP-4742178-A1 - TOOTH LOCALIZATION

EP4742178A1EP 4742178 A1EP4742178 A1EP 4742178A1EP-4742178-A1

Abstract

Proposed concepts aim to provide schemes, solutions, concepts, designs, methods, and systems pertaining to tooth localization and identification in dental imaging. In particular, embodiments aim to provide accurate and robust methods for matching tooth images from routine scans to reference images from calibration scans, enabling reliable tooth identification and localization without requiring a highly accurate universal tooth classification model.

Inventors

  • VLUTTERS, RUUD
  • GALLUCCI, Alessio

Assignees

  • Koninklijke Philips N.V.

Dates

Publication Date
20260513
Application Date
20241112

Claims (15)

  1. A method for tooth localization, the method comprising: obtaining a set of reference images of the subject's teeth, each reference image comprising an image of a respective tooth of the subject at a known tooth location; obtaining a set of scan images of the subject's teeth, the scan images being captured during performance of an oral scanning routine on the subject, and each scan image comprising an image of a respective tooth of the subject; calculating measures of similarity of set of the scan images with the set of reference images; and identifying a tooth of a scan image based on the calculated measures of similarity
  2. The method of claim 1, wherein obtaining a set of reference images comprises: performing a calibration scan of the subject's teeth to obtain a set of reference images of the subject's teeth.
  3. The method of claim 2, wherein the calibration scan is performed with guidance to obtain reference images of each tooth of the subject with known locations.
  4. The method of any of claims 1 to 3, wherein the oral scanning routine is performed without guidance.
  5. The method of any of claims 1 to 4, wherein calculating measures of similarity comprises: using an artificial intelligence, AI, model to map tooth images from the reference images and the scan images to embedding vectors; and calculating similarity scores between the embedding vectors of the reference images and the scan images.
  6. The method of claim 5, further comprising using a Hungarian algorithm to match scan images to reference images based on the calculated similarity scores.
  7. The method of claim 5 or 6, wherein the AI model is trained using a self-supervised method to measure similarity between tooth images.
  8. The method of claim 7, wherein the self-supervised method comprises a Siamese learning procedure with a contrastive loss function.
  9. The method of any of claims 1 to 8, wherein at least one of the set of reference images and the set of scan images acquired by oral care device having an image capture device adapted to capture images of one or more oral features of the subject.
  10. A computer program product for tooth localization, the computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising: obtaining a set of reference images of the subject's teeth, each reference image comprising an image of a respective tooth of the subject at a known tooth location; obtaining a set of scan images of the subject's teeth, the scan images being captured during performance of an oral scanning routine on the subject, and each scan image comprising an image of a respective tooth of the subject; calculating measures of similarity of set of the scan images with the set of reference images; and identifying a tooth of a scan image based on the calculated measures of similarity
  11. A system for tooth localization, the system comprising: an interface configured to obtain a set of reference images of the subject's teeth, each reference image comprising an image of a respective tooth of the subject at a known tooth location, and to obtain a set of scan images of the subject's teeth, the scan images being captured during performance of an oral scanning routine on the subject, and each scan image comprising an image of a respective tooth of the subject; a processor arrangement configured to calculate measures of similarity of set of the scan images with the set of reference images; and a controller configured to generate a signal identifying a tooth of a scan image based on the calculated measures of similarity.
  12. The system of claim 11 wherein the interface comprises an image capture device adapted to capture images of one or more oral features of the subject.
  13. The system of claim 11 or 12, wherein the processor arrangement is configured to use an artificial intelligence, AI, model to map tooth images from the reference images and the scan images to embedding vectors, and to calculate similarity scores between the embedding vectors of the reference images and the scan images.
  14. The system of claim 13, wherein the AI model is trained using a self-supervised method to measure similarity between tooth images
  15. An oral care device comprising: an image capture device adapted to capture images of one or more oral features of a user; and a system according to any of claims 11 to 14, and optionally wherein the oral care device comprises an Intra-Oral Scanner.

Description

FIELD OF THE INVENTION The present invention relates to the field of oral care, and in particular the field of tooth localization in dental imaging. BACKGROUND OF THE INVENTION Dental imaging and analysis have become increasingly important tools in oral healthcare, allowing for early detection and monitoring of various dental conditions. Traditional methods of dental examination rely heavily on visual inspection by dental professionals, which can be subjective and may miss subtle changes or early signs of oral issues. Advanced imaging technologies, such as intraoral cameras/scanners and fluorescence-based systems, have emerged to provide more detailed and objective information about oral health. One significant challenge in the field of dental imaging is the accurate identification and localization of individual teeth within captured images or scans. This task is important for proper diagnosis, treatment planning, and monitoring of dental conditions. However, the wide variety of tooth shapes, sizes, and arrangements among individuals makes it difficult to develop a universal system for tooth identification that works reliably across diverse populations. Furthermore, the dynamic nature of the oral environment, including factors such as tooth movement, tooth colour, relative camera pose and/or position, dental work, changes in teeth condition and/or presence, orthodontics conditions, and changes in oral hygiene, can complicate the consistent identification of teeth across multiple imaging sessions. This poses a particular challenge for longitudinal monitoring of oral health, where it is essential to track changes in specific teeth over time. SUMMARY OF THE INVENTION The invention is defined by the claims. According to an aspect of the invention, there is provided a method for tooth localization. The method comprises: obtaining a set of reference images of the subject's teeth, each reference image comprising an image of a respective tooth of the subject at a known tooth location; obtaining a set of scan images of the subject's teeth, the scan images being captured during performance of an oral scanning routine on the subject, and each scan image comprising an image of a respective tooth of the subject; calculating measures of similarity of set of the scan images with the set of reference images; and identifying a tooth of a scan image based on the calculated measures of similarity. This may, for example, allow for automated and accurate localization and identification of a teeth in dental images. Proposed concepts thus aim to provide schemes, solutions, concepts, designs, methods, and systems pertaining to tooth localization and identification in dental imaging. In particular, embodiments aim to provide accurate and robust methods for matching tooth images from routine scans to reference images from calibration scans, enabling reliable tooth identification and localization without requiring a highly accurate universal tooth classification model. Embodiments leverage the idea that while teeth vary widely across individuals, an individual's teeth remain relatively consistent over time. By using a personalized set of reference images obtained through a guided calibration scan, embodiments may more accurately identify and localize teeth in subsequent routine scans. This may be particularly advantageous for dealing with variations in tooth appearance, dental work, and common orthodontic issues that may pose challenges for traditional tooth classification/identification methods. Furthermore, by incorporating the proposed concept(s) into oral care devices with integrated imaging capabilities, such as Intra-Oral Scanners, embodiments may enable a comprehensive approach to oral health monitoring and personalized care. Such integration may allow for regular, convenient scanning and analysis of teeth, potentially leading to earlier detection of dental issues and more effective preventive care strategies. Overall, the proposals may address a significant challenge in dental imaging and analysis, offering a practical concept that has the potential to improve and/or personalize oral healthcare outcomes for a wide range of individuals. Embodiments may enable accurate tooth identification and localization without requiring a highly accurate universal tooth classification model (which is very difficult to build), making it more robust to variations in tooth appearance across individuals. The method may further comprise performing a calibration scan of the subject's teeth to obtain a set of reference images of the subject's teeth and/or oral surface(s) (e.g. a combination of the teeth (number) and the surface (lingual=inner, facial/buccal=outer, occlusal=chewing)). Performing a calibration scan allows for the creation of a personalized reference set of tooth images for each subject, helping to improve the accuracy of subsequent tooth identification. The calibration scan may be performed with guidance to obtain reference images o