公开(公告)号：WO2021198243A1

公开(公告)日：2021-10-07

申请号：PCT/EP2021/058272

申请日：2021-03-30

Applicant: CARL ZEISS AG

Inventor： FREYTAG, Alexander ,  KUNGEL, Christian ,  EIBL, Matthias ,  KINDT, Johannes

IPC: G06T7/00 ,  G06T11/00 ,  G06T11/001 ,  G06T2207/10056 ,  G06T2207/20084 ,  G06T2207/30024 ,  G06T7/0012

Abstract: Method for virtually staining a tissue sample comprising selecting a virtual stain, obtaining digital imaging data of the tissue sample, wherein the digital imaging data of the tissue sample has been acquired using one or more image modalities, determining a region of interest (ROI) of the tissue sample, providing an output image depicting the tissue sample comprising the virtual stain only in the ROI. Further, it is proposed a device comprising a processor configured to perform the method.

公开(公告)号：WO2021198241A1

公开(公告)日：2021-10-07

申请号：PCT/EP2021/058270

申请日：2021-03-30

Applicant: CARL ZEISS AG

Inventor： FREYTAG, Alexander ,  KUNGEL, Christian

IPC: G06T7/00 ,  G06T11/00 ,  G06T11/001 ,  G06T2207/10056 ,  G06T2207/20084 ,  G06T2207/30024 ,  G06T7/0012

Abstract: A method of virtual staining of a tissue sample includes obtaining multiple sets of imaging data. The imaging data depicts a tissue sample and has been acquired using multiple imaging modalities. Further, the method includes fusing and processing the multiple sets of imaging data in a machine-learning logic. The machine-learning logic is configured to provide at least one output image. Each one of the at least one output image depicts the tissue sample comprising a respective virtual stain.

公开(公告)号：WO2021198279A1

公开(公告)日：2021-10-07

申请号：PCT/EP2021/058335

申请日：2021-03-30

Applicant: CARL ZEISS AG

Inventor： FREYTAG, Alexander ,  KUNGEL, Christian

IPC: G06T7/00 ,  G06K9/62 ,  G06K9/0014 ,  G06T2207/10056 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30096 ,  G06T7/0012

Abstract: Methods for obtaining scores based on imaging data of tissue samples using machine-learning logics, training methods and corresponding devices are provided. With various devices and methods disclosed, automatic scoring of tissue samples becomes possible.

公开(公告)号：WO2021198252A1

公开(公告)日：2021-10-07

申请号：PCT/EP2021/058283

申请日：2021-03-30

Applicant: CARL ZEISS AG

Inventor： FREYTAG, Alexander ,  KUNGEL, Christian

IPC: G06T7/00 ,  G06T11/00 ,  G06T11/001 ,  G06T2207/10056 ,  G06T2207/20084 ,  G06T2207/30024 ,  G06T7/0012

Abstract: It is proposed method for training of a virtual staining logic, wherein the virtual staining logic comprises a cycle generative adversarial network, wherein the cycle generative adversarial network is configured to receive imaging data relating to a tissue sample which has been acquired using a group of image modalities and to provide an output image depicting the tissue sample comprising a virtual stain, wherein the method for training comprises acquiring training imaging data relating to a first plurality of tissue samples using the group of image modalities, obtaining multiple reference images depicting a second plurality of tissue samples, wherein the tissue sample of the second plurality of tissue samples comprises a chemical stain, training of the cycle generative adversarial network logic with the acquired imaging data and the multiple reference images. Furthermore, it is proposed a method for virtually staining. In addition, it is proposed a device for performing the methods.

公开(公告)号：WO2021198244A1

公开(公告)日：2021-10-07

申请号：PCT/EP2021/058273

申请日：2021-03-30

Applicant: CARL ZEISS AG

Inventor： FREYTAG, Alexander ,  KUNGEL, Christian

IPC: G06T7/00 ,  G06T11/00 ,  G06T11/001 ,  G06T2207/10056 ,  G06T2207/20084 ,  G06T2207/30024 ,  G06T7/0012

Abstract: A method of virtual staining of a tissue sample includes obtaining imaging data (501-503) depicting the tissue sample. The method also includes processing the imaging data (501- 503) in at least one machine-learning logic (500), the at least one machine-learning logic (500) being configured to provide multiple output images (521) all comprising a given virtual stain of the tissue sample, the multiple output images (521) depicting the tissue sample comprising the given virtual stain at different colorings associated with different staining laboratory processes. The method further includes obtaining, from the at least one machine- learning logic (500), at least one output image (521) of the multiple output images.

公开(公告)号：WO2021198247A1

公开(公告)日：2021-10-07

申请号：PCT/EP2021/058277

申请日：2021-03-30

Applicant: CARL ZEISS AG

Inventor： FREYTAG, Alexander ,  KUNGEL, Christian

IPC: G06K9/62 ,  G06T7/00 ,  G06T11/00 ,  G06N3/04 ,  G06N3/08 ,  G06N5/00 ,  G06N7/00 ,  G06K9/6259 ,  G06K9/6262 ,  G06K9/6289 ,  G06N3/0454 ,  G06N3/088 ,  G06N5/003 ,  G06N7/005 ,  G06T11/001 ,  G06T2207/10056 ,  G06T2207/10064 ,  G06T2207/10072 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30024 ,  G06T2207/30168 ,  G06T7/0002

Abstract: It is proposed a method for evaluating image modalities for virtual staining of a tissue sample comprising at least one iteration of obtaining multiple sets of training imaging data relating to one or more tissue samples, wherein each set of the multiple sets of imaging data has been acquired using a different image modality of a group of image modalities, obtaining multiple reference images depicting the one or more tissue samples comprising one or more chemical stains, processing the multiple sets of training imaging data in a machine-learning logic, obtaining, from the machine-learning logic and for each one of the one or more training imaging data, training output images comprising one or more virtual stains corresponding to the one or more chemical stains, performing the training of the machine-learning logic by updating parameter values of the machine-learning logic based on a comparison between such reference images and training output images that are associated with corresponding chemical stains and virtual stains, determining a virtual staining accuracy of the trained machine-learning logic for each one of the one or more virtual stains. The evaluating is based on the one or more virtual staining accuracies. Moreover, it is proposed a message for virtual staining and a device for performing at least one of the methods.