Siemens Healthcare Gmbh

Siemens Healthineers Ag

Georg-august-universitaet Goettingen Stiftung Oeffentlichen Rechts Universitaetsmedizin Goettingen

Andre Aichert

Arnaud Arindra Adiyoso

Martin Kraus

Birgi Tamersoy

Andreas Maier

Tobias Heimann

Marvin Teichmann

Hanibal Bohnenberger

Tobias Wuerfl

One or more example embodiments are methods and corresponding systems for providing a training data set for training a segmentation algorithm for segmenting whole-slide images in digital pathology as well as the use of the training data and corresponding ML segmentation algorithms. For example, a first segmentation of a whole slide image is refined based on an automatically generated annotation which has a higher level of detail than the first segmentation. A second segmentation results, which may be used as a ground truth for training the ML segmentation algorithm on the basis of the whole slide image.

Methods and systems for providing training data sets for training a machine-learned segmentation algorithm for use in digital pathology

Various example embodiments pertain to processing images that depict tissue samples using a neural network algorithm. The neural network algorithm includes multiple encoder branches that are copies of each other that share the same parameters. The encoder branches can, accordingly, be referred to as Siamese copies of each other.

Digital pathology using an artificial neural network

A system and a method are provided for a parameter update. In an embodiment, the method includes obtaining, by a first entity, a function and parameter data from a second entity; selecting data samples provided by the first entities; providing a plurality of mutually isolated computing instances; assigning and providing the selected data samples to the computing instances; calculating, within each computing instance, results of the function; calculating averages over the results; determining whether the function fulfils a security criterion, and, if so: providing the calculated average for the gradient of the loss function and/or the calculated average of the output value and/or updated parameter data to the second entity.

Securely performing parameter data updates

A method for determining assignment data is carried out in a method for determining a geometry calibration. The 3D calibration phantom features a calibration object with a number of calibration elements, which are arranged so that a descriptor based on the spatial arrangement is projectively invariant. Based upon the descriptor the calibration elements mapped in the 2D transmission element can be assigned to the calibration elements of the calibration object, so that the geometry calibration is determined on the basis of this assignment and the arrangement of the calibration elements in the three-dimensional space as well as on the 2D image.

Method and device for determining a geometry calibration for an imaging device and also method for determining assignment data for the geometry calibration

Growing community of inventors

Erlangen, Germany

Andre Aichert