US 11,989,959 B2
Method and computer program for clustering large multiplexed spatially resolved data of a biological sample
Lucas Pelkmans, Zürich (CH); and Gabriele Gut, Münchenstein (CH)
Appl. No. 17/257,959
Filed by UNIVERSITÄT ZÜRICH, Zürich (CH)
PCT Filed Jul. 5, 2019, PCT No. PCT/EP2019/068181
§ 371(c)(1), (2) Date Jan. 5, 2021,
PCT Pub. No. WO2020/008071, PCT Pub. Date Jan. 9, 2020.
Claims priority of application No. 18182243 (EP), filed on Jul. 6, 2018; and application No. 19154219 (EP), filed on Jan. 29, 2019.
Prior Publication US 2021/0224510 A1, Jul. 22, 2021
Int. Cl. G06V 20/69 (2022.01); G01N 1/30 (2006.01); G06F 18/2137 (2023.01); G06F 18/2323 (2023.01); G06V 10/762 (2022.01)
CPC G06V 20/695 (2022.01) [G01N 1/30 (2013.01); G06F 18/2137 (2023.01); G06F 18/2323 (2023.01); G06V 10/7635 (2022.01)] 16 Claims
OG exemplary drawing
 
1. A method for processing multiplexed image data of a biological sample, the method comprising the steps of:
a) Recording a plurality of images of a biological sample, wherein the plurality of images comprises images having a different entity of the biological sample targeted with a predefined stain;
b) Registering the plurality of images, thereby determining spatially corresponding image pixels;
c) Associating the spatially corresponding image pixels to a pixel profile, wherein each pixel profile comprises the pixel values of the spatially corresponding pixels and wherein the pixel profile is associated with the respective image coordinate of the spatially corresponding pixels;
d) Pooling the pixel profiles by means of a clustering method configured to determine pixel profiles with similar values, and thereby generating a plurality of clusters, each comprising pixel profiles with similar pixel values;
e) For each cluster assigning a cluster value to the image coordinate of the pixel profiles comprised by said cluster and thereby generating a cluster image with cluster pixels, wherein
i) For each cluster pixel in the cluster image the cluster values of adjacent cluster pixels in the cluster image are determined and thereby for each cluster value pair a probability of adjacency is determined;
ii) Generating at least one randomized cluster image, wherein the image coordinates of the cluster pixels in the at least one randomized cluster image are randomly inter-changed;
iii) For each cluster pixel in the at least one randomized cluster image the cluster values of adjacent cluster pixels in the randomized cluster image are determined and thereby for each cluster value pair a probability of random adjacency is determined and wherein the probability of random adjacency is determined for all of the randomized cluster images;
Determining an adjusted probability of adjacency by calculating a deviation, particularly a difference between the probability of adjacency and the probability of random adjacency.