Abstract: Systems and methods for training and implementing a machine learning algorithm to generate feature maps depicting spatial patterns of features associated with osteolysis, synovitis, or both. MRI data, including multispectral imaging data, are input to the trained machine learning algorithm to generate the feature maps, which may indicate features such as a location and probability of a pathology classification, a severity of synovitis, a type of synovitis, a synovial membrane thickness, and other features associated with osteolysis or synovitis. In some implementations, synovial anatomy are segmented in the MRI data before inputting the MRI data to the machine learning algorithm. These segmented MRI data may be generated using another trained machine learning algorithm.

Abstract: Materials-handling vehicles, such as counterbalanced lift trucks, can incorporate a counterweight that includes a frame, a cavity, and a sensor-mounting recess. The counterweight may be configured to provide an unobstructed horizontal line of sight of at least 180 degrees for a sensor, such as an object-detection sensor, that is mounted within the sensor-mounting recess. A lift truck can include a lift assembly comprising a mast and a pair of forks, an operator compartment comprising truck steering and speed controls, a plurality of wheels, an energy source, a counterweight comprising a frame and a sensor-mounting recess, and a sensor.

Abstract: A developer tool is used to programmatically calculate the contrast between the text and background in a user interface. A luminosity ratio is automatically calculated by the tool by capturing an image of a user interface, archiving user interface elements that do not contain text, performing optical character recognition on the remaining user interface elements to find pixels on the text part of the user interface element and pixels on the background part of the user interface element, and finally calculating the luminosity ratio. Results including errors, warnings and luminosity ratio for one or more hotspots in the image can be returned. Results can be used to identify user interface elements that comply or fail to comply with visibility guidelines for contrast between user interface text and background.

Abstract: A developer tool is used to programmatically calculate the contrast between the text and background in a user interface. A luminosity ratio is automatically calculated by the tool by capturing an image of a user interface, archiving user interface elements that do not contain text, performing optical character recognition on the remaining user interface elements to find pixels on the text part of the user interface element and pixels on the background part of the user interface element, and finally calculating the luminosity ratio. Results including errors, warnings and luminosity ratio for one or more hotspots in the image can be returned. Results can be used to identify user interface elements that comply or fail to comply with visibility guidelines for contrast between user interface text and background.

Abstract: Various technologies and techniques are disclosed that improve the operation of accessibility applications. A graphics pipe is provided that can be called in user mode from multiple accessibility programs. A request is received from an accessibility application to access the graphics pipe, and a connection is established. The accessibility application listens to the graphics pipe for particular content of interest and builds a model based on that content. The model is used to deliver content to an end user appropriately. Screen captures can be performed on at least part of the content and then rendered onto another surface.

Abstract: Described is an engine and API that couples an application program to an effects program such as a magnification program. For example, source content may be magnified for viewing in an output region. Magnification may be accomplished by identifying a magnification window to the magnification engine, a source region to magnify, a magnification transform, and possibly filtering criteria, such as any windows to include or exclude from magnification. A request to display a region of displayed graphics as modified by a transform may be received, and when graphics commands corresponding to that region are received, the graphics commands are processed to show a transformed representation of the region. The engine and API may work with immediate mode graphics primitives (e.g., GDI commands) and retained mode graphics primitives (e.g., primitives corresponding to a rendering tree; a composition engine composes the output, including any magnified output.