Detection of tooth condition using reflectance images with red and green fluorescence

Patent metadata

Field	Value
Publication number	US-2016125601-A1
Application number	US-201514933294-A
Country	US
Kind code	A1
Filing date	Nov 5, 2015
Priority date	Nov 5, 2014
Publication date	May 5, 2016
Grant date	—

Abstract

A system and method for imaging a tooth. The method illuminates the tooth and acquires reflectance image data and illuminates the tooth and acquires fluorescence image data from the tooth. The acquired reflectance and fluorescence image data for the tooth are aligned to form aligned reflectance and fluorescence image data. For one or more pixels of the aligned reflectance and fluorescence image data, at least one feature vector is generated, having data derived from one or both of the aligned reflectance and fluorescence image data. The aligned reflectance and fluorescence image data and the at least one feature vector are processed using one or more trained classifiers obtained from a memory that is in signal communication with the computer. Processing results indicative of tooth condition are displayed.

First claim

What is claimed is:

1 . A method for imaging a tooth, the method executed at least in part by a computer, comprising:
illuminating the tooth and acquiring reflectance image data of the tooth; illuminating the tooth and acquiring fluorescence image data from the tooth; aligning the acquired reflectance and fluorescence image data for the tooth to form aligned reflectance and fluorescence image data; for one or more pixels of the aligned reflectance and fluorescence image data, generating at least one feature vector having data derived from one or both of the aligned reflectance and fluorescence image data; processing the aligned reflectance and fluorescence image data and the at least one feature vector using one or more trained classifiers obtained from a memory that is in signal communication with the computer to generate processed data; and displaying, storing, or transmitting the processed data.

2 . The method of claim 1 wherein the at least one feature vector is a multivalued or n-dimensioned data structure, and wherein the at least one feature vector comprises data derived from both the aligned reflectance and fluorescence data.

3 . The method of claim 1 further comprising displaying an image that is representative of the acquired reflectance or fluorescence image data.

4 . The method of claim 1 wherein the processing results comprise a likelihood mapping indicative of tooth condition, wherein the processing results further comprise a composite image generated from the likelihood mapping and the acquired reflectance or fluorescence image data.

5 . The method of claim 1 wherein generating the at least one feature vector comprises transforming the fluorescence image data from tristimulus to polar color data form.

6 . The method of claim 1 wherein the fluorescence image data includes green light in the 490-560 nm range and includes red light with wavelength greater than 600 nm, and wherein generating the at least one feature vector comprises transforming the fluorescence image data from red-green-blue color data to hue-saturation-value color data.

7 . The method of claim 1 wherein processing the aligned reflectance and fluorescence image data and the at least one feature vector comprises using a one-dimensional or two-dimensional linear classifier or Gaussian mixture model.

8 . The method of claim 1 wherein aligning the acquired reflectance and fluorescence data for the tooth comprises:
down-sampling the reflectance and fluorescence image data;
matching image content for at least a first rectangular block in the down-sampled reflectance image data with at least a second rectangular block in the down-sampled fluorescence image data; and
generating a transformation between the reflectance and fluorescence image data according to the match of image content.

9 . A method for imaging a tooth, the method executed at least in part by a computer, comprising:
acquiring reflectance image data of the tooth; acquiring fluorescence image data from the tooth; aligning the acquired reflectance and fluorescence data for the tooth; identifying at least one feature from the acquired reflectance and fluorescence data and forming a composite image using the acquired reflectance and fluorescence image data and the identified feature; and displaying an operator interface that identifies a plurality of teeth of the patient and, for at least one of the identified teeth, allows operator selection of one or more of the reflectance image data, fluorescence image data, and composite image of the tooth for display.

10 . The method of claim 9 wherein identifying at least one feature comprises:
generating at least one feature vector having data derived from the aligned reflectance and/or fluorescence image data; and
processing the aligned reflectance and fluorescence image data and the at least one feature vector using one or more trained classifiers to generate results indicative of tooth condition.

11 . The method of claim 9 further comprising responding to an operator-entered instruction by displaying, for an identified tooth, one or more of the reflectance image data, fluorescence image data, and composite image of the tooth.

12 . The method of claim 9 further comprising displaying a likelihood mapping that shows likelihood of a tooth condition for each of a plurality of teeth.

13 . The method of claim 9 further comprising displaying a chart showing one or more teeth for operator identification, wherein the operator selection is made on a camera that is in signal communication with the computer.

14 . An interactive dental charting system that displays a plurality of teeth for a patient and, for each displayed tooth, provides a display that enables operator selection of both acquired and processed images for the tooth, wherein the acquired images for each tooth include both a reflectance image and a fluorescence image obtained from an intra-oral camera and wherein the processed images for the tooth include a composite image that is indicative of tooth condition.

15 . The system of claim 14 wherein the processed images for the tooth include a likelihood mapping that indicates probability of infection for the selected tooth.

16 . The system of claim 14 wherein the processed images for the tooth include one or more types of images that identify incipient caries, include one or more types of images that identify advanced caries, include one or more types of images that identify dentinal caries, include one or more types of images that identify plaque, and wherein the operator selection includes one, two, three or all of the incipient caries images, the incipient caries images, the dentinal caries images and the plaque images.

17 . The system of claim 16 wherein the dental charting system further enables operator selection of a composite image indicative of tooth condition.

18 . The system of claim 17 wherein the display lists the available types of acquired and processed images for each of a plurality of teeth of the patient.

19 . An intra-oral imaging apparatus, comprising:
an intra-oral camera configured to acquire both reflectance and fluorescence images of a patient's teeth; a control logic processor in signal communication with the intra-oral camera and programmed with instructions to obtain and process the reflectance and fluorescence image data and to distinguish incipient caries from advanced and dentinal caries; and a display in signal communication with the control logic processor for displaying the acquired and processed images and identifying incipient, advanced, and dentinal caries for one or more of the teeth according to processing results.

20 . The apparatus of claim 19 wherein the display further displays a likelihood mapping for infection of one or more of the teeth, wherein the processor distinguishes caries that does not extend through the tooth enamel from caries that extends through the tooth enamel according to the reflectance and fluorescence image data, wherein the display shows the types of acquired and processed images that can be displayed for one or more teeth, wherein the instructions for processing the image data generate one or more feature vectors, and wherein the control logic processor is part of the intra-oral camera.

Assignees

• Carestream Health Inc

Inventors

• Wu Yingqian
• Wang Wei
• Wong Victor C
• Zhang Yan

Classifications

• G06T2207/10024

  Color image · CPC title

• G06T2207/30036

  Dental; Teeth · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• G06T2207/10064

  Fluorescence image · CPC title

• H04N23/555

  for picking-up images in sites, inaccessible due to their dimensions or hazardous conditions, e.g. endoscopes or borescopes · CPC title