METHOD FOR ASSESSING IMAGE FOCUS QUALITY

• US 20080137938A1
• Filed: 12/11/2006
• Published: 06/12/2008
• Est. Priority Date: 12/11/2006
• Status: Active Grant

• Current Assignee: Cytyc Corporation (Hologic Incorporated)
• Original Assignee: Cytyc Corporation (Hologic Incorporated)
• Inventors: Zahniser, Michael
• Granted Patent: US 7,769,219 B2
• Time in Patent Office: Days
• US Class Current: 382/133
• CPC Class Codes: G02B 21/244 using image analysis techniques
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection

Abstract

A method for determining the quality of focus of a digital image of a biological specimen includes obtaining a digital image of a specimen using a specimen imaging apparatus. A measure of image texture is calculated at two different scales, and the measurements are compared to determine how much high-resolution data the image contains compared to low-resolution data. The texture measurement may, for example, be a Brenner auto-focus score calculated from the means of adjacent pairs of pixels for the high-resolution measurement and from the means of adjacent triples of pixels for the low-resolution measurement. A score indicative of the quality of focus is then established based on a function of the low-resolution and high-resolution measurements. This score may be used by an automated imaging device to verify that image quality is acceptable. The device may adjust the focus and acquire new images to replace any that are deemed unacceptable.

Claims

1. A method for determining the quality of focus of a digital image of a biological specimen comprising:

(a) obtaining a digital image of the specimen using the specimen imaging apparatus, the digital image comprising a plurality of pixels;

(b) calculating a measure of small-scale texture in the image;

(c) calculating a measure of larger-scale texture in the image, and

(d) establishing an estimate of the image focus quality based on a comparison of the small-scale and larger-scale texture measurements.

2. The method of claim 1, wherein the small-scale texture is measured by squaring the differences of the means of adjacent pairs of pixels across the entire digital image.

3. The method of claim 1, wherein the larger-scale texture is measured by squaring the differences of the means of adjacent triples of pixels across the entire digital image.

4. The method of claim 1, wherein a function of the small-scale and larger-scale texture measurements is matched to an experimentally derived data set to determine an estimated displacement from the ideal focal height.

5. The method of claim 4, wherein a best-fit equation is calculated for the experimentally derived data set and its inverse is used to map measured scores to predicted displacements.

6. The method of claim 4, wherein the function applied to the small-scale and larger-scale texture measurements is to take their quotient.

7. The method of claim 5, wherein a bell-shaped curve is fit to the experimentally derived data.

8. The method of claim 1, wherein an automated imaging device obtains a plurality of digital images of different locations within a specimen and the focus quality is estimated for at least some of the images.

9. The method of claim 8, wherein the images are accepted or rejected based on the estimated focus quality.

10. The method of claim 9, wherein rejection of an image causes acquisition of a new image at the same location but at a different focal plane.

11. The method of claim 9, further comprising the step of saving the locations of all rejected images and re-acquiring images at those locations once the device has finished scanning the remainder of the sample.

12. The method of claim 9, wherein the number of rejected images is monitored for a given sample and if that number crosses a certain threshold, the imaging device will reprocess the entire sample.

13. The method of claim 8, further comprising the step of adjusting the focal height based at least in part of the estimated quality and acquiring subsequent images.

14. The method of claim 13, wherein the imaging device adds an offset to the height at which it would otherwise have acquired each image and adjusts this offset up or down depending on whether the previous adjustments resulted in an increase or decrease in focus quality.

15. The method of claim 8, wherein the automated imaging device monitors focus quality estimates and identifies hardware malfunctions based on the monitoring.

16. The method of claim 15, wherein a suspected hardware malfunction causes the imaging device to flag the results from the samples processed.

17. The method of claim 15, further comprising the step of alerting a user to the suspected malfunction.

18. The method of claim 4, further comprising the step of adjusting the focal height based at least in part of the estimated displacement value and acquiring subsequent images.

19. The method of claim 18, wherein the step is repeated until the focal height converges on the ideal focal height.

20. A method of focusing on a specimen using an automatic imaging apparatus, the method comprising:

(a) obtaining three digital images of the specimen at different focal heights using the imaging apparatus, the digital image comprising a plurality of pixels;

(b) calculating a Brenner auto-focus score for each of the three digital images;

(c) fitting an exponentiated quadratic function to the (x, y) points formed by the three focal heights and the corresponding image Brenner scores;

(d) moving to a new focal height based on the function of step (c), acquiring a new image, and calculating the Brenner score for the new image, and

(e) repeating step (d) until an ideal focal height is reached.

21. The method of claim 20, wherein each new image is acquired at the peak of the fit function calculated from the previous data points.

22. The method of claim 21, wherein the ideal focal height is reached when fit function height, estimated when incorporating the newest data point into the fit, is within a certain threshold of the previous function height estimate.

23. The method of claim 20, wherein the fit function is calculated using only three data points:

(a) The data point with the highest score found so far;

(b) The data point with the focal height that is closest to point (a), but below it; and

(c) The data point with the focal height that is closest to point (a), but above it.

24. The method of claim 23, wherein with each step the focal height is adjusted to the height which is as close as possible to the peak of the fit function while remaining at least a certain distance away from all focal heights checked so far.

25. The method of claim 23, wherein the ideal focal height is reached when the three points used in the fit are all within a certain distance of each other.

26. A method of focusing a specimen imaging apparatus comprising:

(a) obtaining a digital image of a specimen using the specimen imaging apparatus, the digital image comprising a plurality of pixels;

(b) calculating a high-resolution Brenner score by squaring the difference of the means of neighboring pairs of pixels and summing the squared differences over a set of pixels within the digital image;

(c) calculating a low-resolution Brenner score by squaring the difference of the means of neighboring triples of pixels and summing the squared differences over a set of pixels within the digital image;

(d) calculating a ratio comprising the high-resolution Brenner score divided by the low-resolution Brenner score; and

(e) adjusting the focal length of the specimen imaging apparatus based on the ratio.

27. The method of claim 26, further comprising the step of obtaining a digital image of the specimen at the adjusted focal length and repeating steps (b) through (e).

28. The method of claim 27, wherein the steps (b) through (e) are repeated a plurality of times with different images at the adjusted focal length.