Digital microscope slide scanning system and methods

摘要

Provided herein are systems methods including a design of a microscope slide scanner for digital pathology applications which provides high quality images and automated batch-mode operation at low cost. The instrument architecture is advantageously based on a convergence of high performance, yet low cost, computing technologies, interfaces and software standards to enable high quality digital microscopy at very low cost. Also provided is a method based in part on a stitching method that allows for dividing an image into a number of overlapping tiles and reconstituting the image with a magnification without substantial loss of accuracy. A scanner is employed in capturing snapshot images. The method allows for overlapping images captured in consecutive snapshots.

主权项

1. A method for reconstituting an image comprising dividing an image into a number of overlapping tiles captured in consecutive snapshots; reconstituting the image with a magnification without substantial loss of accuracy; and displaying or storing a reconstituted image; wherein the method comprises calculating stitch points between two consecutive snapshots acquired by the scanner for stitching adjacent tiles, wherein the stitching involves an input comprising a set of two images and wherein an output comprising stitch points for stitching the two images and further comprises:
i. converting input image to a gray scale image; ii. in the overlapped region of first image, finding scan boxes with good contrast; iii. sorting scan boxes with descending order of contrast so as to start a search for the stitch point with scan box of highest gradient; iv. selecting a scan box with good contrast in first image and start search for a matching scan box in the second image; v. for all the good contrast scan boxes in the first image, repeat following steps, till a stitch point is determined:
a. perform first level cross correlation with the scan box of size 3×3 with an overlap region in the second image;b. if the value of cross correlation of this box is greater than 0.91, extend the scan box to the size of 16×16 and perform a second level cross correlation with the corresponding extended box in the second image;c. find the highest value of the second level cross correlation and if it is less than 0.9, get the next scan box in the list and go to step b above;d. make a list of all the scan boxes in the second image whose cross correlation with the scan box in first image is less than the highest value of cross correlation by 0.05;e. for the list of scan boxes, find a diagonally opposite scan box and perform cross correlation with a diagonally opposite scan box in the first image;f. find the highest value of cross correlation of diagonally opposite scan boxes in the left image with the diagonally opposite scan boxes in the right image;g. if this value is less than 0.455, get the next scan box in the list and go to step b above; andh. otherwise break the loop and declare the stitch point found; whereby the left top coordinate of the scan box in the first image and left top coordinate of the scan box with highest correlation in the second image are declared as the stitch points.