Method of imaging multiple retinal structures

Patent metadata

Field	Value
Publication number	US-10092181-B2
Application number	US-201615097040-A
Country	US
Kind code	B2
Filing date	Apr 12, 2016
Priority date	Apr 29, 2015
Publication date	Oct 9, 2018
Grant date	Oct 9, 2018

Abstract

Methods and apparatus for imaging multiple retinal structures and montaging of multiple retinal images are provided. The method involves cross-correlating images from different imaging channels to compensate for intra-frame distortion due to retinal movement during image acquisition, and conducting a second cross-correlation to filter out any motion artifacts in the images. The resultant images are combined to generate a composite image. The method also involves controlling light directed on the retina spatially and temporally.

First claim

The invention claimed is:

1. A method of imaging multiple retinal structures comprising:
direct light on a portion of the retina to achieve irradiance;
acquire a first image frame and register successive image frames of a portion of the retina from a first imaging channel, and simultaneously acquire and co-register second image frames of the portion of the retina from a second, different imaging channel;
divide the first image frame and each successive image frame into corresponding multiple strips, and cross-correlate each successive image frame strip with a corresponding first image frame strip to compensate for intra-frame distortion due to retinal movement during image acquisition, and apply these strip motions to the second image frames and co-register the second image strips;
re-divide the registered successive image frames into second strips, and conduct a second cross-correlation of each corresponding second strip, thereby filtering out any motion artifacts in the registered images and register the first images, and
combine the final first images and generate a composite image for the first imaging channel, and co-register and combine final second images and generate a composite image for the second imaging channel.

2. The method of claim 1 , further comprising simultaneously acquiring and co-registering third image frames of a portion of the retina from a third, different imaging channel; and combining final third images and generating a composite image thereon.

3. The method of claim 1 , further comprising controlling timing and intensity of light directed on the portion of the retina, such that light exposure on said portion is minimized.

4. The method of claim 1 , wherein the light is directed with a light source including a shutter, and the method further comprises controlling the shutter so that light is directed on the portion of the retina only during acquisition of images thereof.

5. The method of claim 1 , wherein the light source includes an acousto-optic modulator, and the method further comprises modulating the power of light directed on the portion of the retina, thereby achieving uniform retinal irradiance.

6. The method of claim 1 , wherein images are acquired and registered with a scanning light microscope or scanning light ophthalmoscope.

7. The method of claim 6 , wherein images are acquired and registered with an adaptive optics scanning light ophthalmoscope including a wavefront corrector.

8. The method of claim 1 , wherein the first channel wavelength and the second channel wavelength are different.

9. The method of claim 1 , wherein the irradiance includes at least one of fluorescence, reflectance and scattering.

10. The method of claim 1 , wherein the second correlation is initiated while additional second images are still being acquired.

11. The method of claim 1 , wherein the second correlation is initiated after second images are acquired.

12. The method of claim 1 , wherein the composite images are displayed on a display device in real-time.

13. The method of claim 1 , wherein the composite image is a high signal-to-noise ratio image of structures from multiple retinal layers of a living animal.

14. The method of claim 1 , wherein the images of structures from multiple retinal layers of a living animal are obtained concurrently.

15. The method of claim 1 , further comprising comparing the cross-correlated first image strips with a threshold, and registering first image strips meeting the threshold.

16. The method of claim 1 , wherein several second final composite images generated from co-registration separated by short temporal intervals are registered and combined to compensate for temporal variation in a lateral position offset between the first imaging channel and second imaging channel.

17. The method of claim 1 , wherein a lateral position offset between images of the first and second imaging channels is recovered by tracking pupil position.

Assignees

• Yang Qiang
• Rossi Ethan Andrew
• Univ Rochester

Inventors

• Yang Qiang
• Rossi Ethan Andrew

Classifications

• G02B27/0031

  for scanning purposes · CPC title

• A61B3/14

  Arrangements specially adapted for eye photography · CPC title

• A61B3/0058

  for multiple images · CPC title

• A61B3/0025

  characterised by electronic signal processing, e.g. eye models · CPC title

• G02B27/0068

  having means for controlling the degree of correction, e.g. using phase modulators, movable elements (controlling the phase of light using moving or deformable elements G02B26/06) · CPC title