METHOD FOR OPERATING A DISTRIBUTED DIGITAL MICROSCOPY SYSTEM AND DISTRIBUTED DIGITAL MICROSCOPY SYSTEM

    公开(公告)号:EP4375883A1

    公开(公告)日:2024-05-29

    申请号:EP22209758.6

    申请日:2022-11-26

    Inventor: Wildner, Ludwig

    Abstract: A method for operating a distributed digital microscopy system (100) is provided. The distributed digital microscopy system (100) comprises a user interface system (82), comprising a user computer (80) and a digital microscope (2) having an overview image acquisition portion (60) and a microscopic image acquisition portion (20), and a server system (90), remote from the user interface system (82). The method comprises: with the overview image acquisition portion (60) of the digital microscope (2), generating a first group of overview images for a first group of microscope slides; segmenting each of the first group of overview images with a pre-trained artificial intelligence, having been trained with an initial annotated training set of overview images; for a subset of the first group of overview images, receiving user input, representing updated segmented versions of the subset of the first group of overview images, via the user computer (80); at the server system (90), providing an updated annotated training set of overview images by adding additional annotated overview images, based on said subset of the first group of overview images and said updated segmented versions of the subset of the first group of overview images, to the initial annotated training set of overview images; at the server system (90), generating a re-trained artificial intelligence, trained with said updated annotated training set of overview images; with the overview image acquisition portion (60) of the digital microscope (2), generating a second group of overview images for a second group of microscope slides; segmenting each of the second group of overview images with the re-trained artificial intelligence; and with the microscopic image acquisition portion (20) of the digital microscope (2), selectively scanning those regions of the second group of microscope slides that correspond to one or more predetermined segmentation classes.

    MICROSCOPE-BASED SUPER-RESOLUTION METHOD AND APPARATUS, DEVICE AND MEDIUM

    公开(公告)号:EP4365774A1

    公开(公告)日:2024-05-08

    申请号:EP22836674.6

    申请日:2022-06-13

    Inventor: CAI, De HAN, Xiao

    Abstract: The present disclosure discloses a microscope-based super-resolution method and apparatus, device and medium, and relates to the field of image processing. The method includes: acquiring an input image and at least one auxiliary image, an overlap area existing between a display area of the auxiliary image and a target area in the input image, and the input image and the auxiliary image being both images of a first resolution degree (302); registering the input image and the auxiliary image to obtain a registered image (304); acquiring a high-resolution feature based on the registered image, the high-resolution feature configured to represent image feature of a second resolution degree in the target area (306); and reconstructing the high-resolution feature to obtain a target image of the second resolution degree (308). The above-mentioned method is used for processing a microscope image sequence to acquire and reconstruct a high-resolution image good in effect.

    IMAGE GENERATION METHOD
    29.
    发明公开

    公开(公告)号:EP4418199A1

    公开(公告)日:2024-08-21

    申请号:EP24153896.6

    申请日:2024-01-25

    Inventor: YASUDA, Takuya

    Abstract: An image generation method generates a plurality of reduced partial images by reducing a plurality of partial images. The reduced partial images are then combined into a reduced composite image. Then, a reduced observation region corresponding to the observation region is detected from the reduced composite image. Then, an observation region that is not reduced is calculated in accordance with the reduced observation region. Thereafter, among the partial images, those that correspond to the observation region are combined into the observation image. This reduces the storage capacity of the memory as compared to the case where the observation region is detected from the composite image that is not reduced. The storage capacity of the memory is also reduced as compared to the case where the composite image of the entire imaging region is generated.

Patent Agency Ranking