Abstract: An image processing component (IPC) and related method for image registration of at least two sets of image objects. The image objects may be previously derived from a segmentation of imagery of biological material. Based on features of the image objects, an initial global transformation is computed. Using this initial global transformation and its inverse, one or more cross-links between sets of image objects across image objects in the two sets are established. A plurality of locally adapted transformations are then computed for cross-linked objects from the two sets. A combiner (?) then combines the locally adapted transformations t\ into a new registration transform TNEW.

Abstract: The present invention relates to digital pathology. In order to improve the workflow in the process of selecting a region of interest of an unstained sample to be removed for molecular diagnostic, a method (100) is provided for selecting a sample removing area of an unstained sample to be removed for molecular diagnostic. The method comprises the following steps: In a first step 102, also referred to as step a), a reference removing area is selected in a reference image of a reference slice of an object, wherein biological material in the reference slice is stained. In a second step 104, also referred to as step b), a digital sample image of a sample slice of the object is obtained under an imaging setting. The biological material in the sample slice is unstained. The sample slice is received on a sample slide and positioned in an optical path between a light source and an image detector.

Abstract: The present invention relates to digital pathology. In order to improve the workflow in the process of selecting a region of interest of an unstained sample to be removed for molecular diagnostic, a method (100) is provided for selecting a sample removing area of an unstained sample to be removed for molecular diagnostic. The method comprises the following steps: In a first step 102, also referred to as step a), a reference removing area is selected in a reference image of a reference slice of an object, wherein biological material in the reference slice is stained. In a second step 104, also referred to as step b), a digital sample image of a sample slice of the object is obtained under an imaging setting. The biological material in the sample slice is unstained. The sample slice is received on a sample slide and positioned in an optical path between a light source and an image detector.

Abstract: An image processing component (IPC) and related method for image registration of at least two sets of image objects. The image objects may be previously derived from a segmentation of imagery of biological material. Based on features of the image objects, an initial global transformation is computed. Using this initial global transformation and its inverse, one or more cross-links between sets of image objects across image objects in the two sets are established. A plurality of locally adapted transformations are then computed for cross-linked objects from the two sets. A combiner (?) then combines the locally adapted transformations t\ into a new registration transform TNEW.