Leica Biosystems Imaging, Inc.

An apparatus for interacting with one or more digital whole slide images (WSIs) acquired by an imaging device includes a display, a memory, and one or more hardware processors. The memory is configured to store computer-executable instructions. The one or more hardware processors are in communication with the display and the memory. The one or more hardware processors are configured to drive the display using the computer-executable instructions of the memory such that the one or more hardware processors are configured to generate a user interface. The user interface includes an image layer and one or more mask layers overlaying the image layer. The one or more mask layers have a predetermined opacity level. The one or more hardware processors are further configured to automatically adjust the predetermined opacity level as a function of a magnification level of the image layer.

Features are disclosed for rendering an image using a GPU and CPU based rendering pipeline. An imaging system may include a GPU and a CPU that each include a portion of an image renderer. A component of the GPU can process the set of image tiles to generate a texture with a wrapped tile coordinate. A component of the GPU can further store the image data as a level of a texture pyramid. As subsequent image data is received, a component of the GPU can access the texture pyramid to determine previously stored image tiles. A component of the GPU can use the previously stored image tiles in rendering the subsequent image data.

Control processes for a slide-scanning system comprising a carousel with a plurality of rack slots configured to receive slide racks via an exposed portion of the scanning system. In an embodiment, initializing the scanning system comprises automatically homing back-end and front-end components, wherein the front-end components comprise the carousel. An inventory of all slide racks in the carousel is automatically generated. If any slide rack was being processed by any back-end components, the slide rack is automatically unloaded into a corresponding rack slot. In addition, the carousel is automatically positioned to expose a starting subset of the rack slots within the exposed portion. This starting subset may comprise a maximum segment of adjacent empty rack slots.

A stall detection system is provided that determines if a glass slide in motion during an automated process is at risk for being damaged and stops motion in the event of an unacceptable risk of damage. The system includes one or more motors configured to move a glass slide (directly or indirectly). The motors are configured to generate a load resistance value. The system includes one or more processors that monitor the load resistance value of a motor during motion and compares the load resistance value to a predetermined threshold resistance value to determine the risk of a glass slide being damaged. The predetermined threshold resistance value may correspond to a risk of slide breakage in response to a force applied to any surface of the slide, or a risk of losing a controlling grip on a slide rack, or a risk of a motor skipping a motor step.

Features are disclosed for imaging block and slice samples using an imaging system. The imaging system can link the images by identifiers associated with the block and slice samples. The imaging system can train a machine learning algorithm based on correctly linked images. In some embodiments, the trained machine learning algorithm may include an image analysis module or a convolutional neural network. The imaging system can use the trained machine learning algorithm in order to determine a confidence score of a match between the block and the slice samples. The trained machine learning algorithm can use features of the block and the slice samples such as shape and tissue morphology to determine whether the samples match. In some embodiments, when the confidence score is below a certain threshold, the imaging system can alert a user that the samples may not match.

Real-time focusing in a slide-scanning system. In an embodiment, focus points are added to an initialized focus map while acquiring a plurality of image stripes of a sample on a glass slide. For each image stripe, a plurality of frames, collectively representing the image stripe, may be acquired using both an imaging line-scan camera and a tilted focusing line-scan camera. Focus points, representing positions of best focus for trusted frames, are added to the focus map. Outlying focus points are removed from the focus map. In some cases, one or more image stripes may be reacquired. Finally, the image stripes are assembled into a composite image of the sample.

Features are disclosed for rendering an image using a GPU and CPU based rendering pipeline. An imaging system may include a GPU and a CPU that each include a portion of an image renderer. A component of the GPU can process the set of image tiles to generate a texture with a wrapped tile coordinate. A component of the GPU can further store the image data as a level of a texture pyramid. As subsequent image data is received, a component of the GPU can access the texture pyramid to determine previously stored image tiles. A component of the GPU can use the previously stored image tiles in rendering the subsequent image data.

An apparatus and methods for determining features in an image of a Hematoxylin and Eosin (H&E) stained tissue sample. An apparatus can be configured to apply a machine learning model to the H&E stained tissue sample image to determine features in the image. Applying the machine learning model can include performing convolution operations on the H&E stained tissue sample image to generate a plurality of initial feature maps, applying octave-convolution-first-layer operations on the initial feature maps to generate initial high-frequency feature maps and low-frequency feature maps, applying octave-convolution operations on the high-frequency initial feature maps to generate refined high-frequency feature maps, applying octave-convolution operations on the low-frequency initial feature maps to generate refined low-frequency feature maps, and applying octave-convolution last layer operations on the refined high-frequency feature maps and the refined low-frequency feature maps to produce combined feature maps.

Systems and methods for similarity learning in digital pathology are provided. In one aspect, an apparatus for generating training image data includes a hardware memory configured to store executable instructions and a hardware processor in communication with the hardware memory, wherein the executable instructions, when executed by the processor, cause the processor to obtain a plurality of histopathology images, classify two or more of the histopathology images as similar or dissimilar, and create a dataset of training image data including the classified histopathology images.

A stall detection system is provided that determines if a glass slide in motion during an automated process is at risk for being damaged and stops motion in the event of an unacceptable risk of damage. The system includes one or more motors configured to move a glass slide (directly or indirectly). The motors are configured to generate a load resistance value. The system includes one or more processors that monitor the load resistance value of a motor during motion and compares the load resistance value to a predetermined threshold resistance value to determine the risk of a glass slide being damaged. The predetermined threshold resistance value may correspond to a risk of slide breakage in response to a force applied to any surface of the slide, or a risk of losing a controlling grip on a slide rack, or a risk of a motor skipping a motor step.