C++ Framework

The MISA++ (Modular image stack analysis for C++) framework allows the creation of C++ applications to process data. It features functions that simplify following common requirements of such applications:

• Automated parallelization of tasks
• Memory-efficient caching of data on hard-drive
• Providing additional algorithm parameters
• Modularization of common tasks
• Assigning quantitative results to data
• Integration into other tools and pipelines

MISA++ is used to develop modules that either provide additional data types and other utility functions (utility modules) or modules that are compiled into an application that solves a specific task (worker modules).

We also provide additional modules that integrate the open source imaging libraries OpenCV and OME Files into our framework.

All features

Core library

Creation of parallelizeable workloads

• Easy creation of parallelizable and atomic tasks
• Easy structuring of tasks into groups
• Dependency management between workloads
• No need for manual thread creation due to automated parallelization

Standardized input parameters

• Highly structured and well standardized way to create algorithm parameters
• Different parameter namespaces
• • Algorithm-specific parameters
• • Sample-specific parameters
• • Global parameters

Memory-efficient handling of large data sets

• Data-access via thread-safe caches
• Easy importing of input data
• Easy creation of output data
• Methods to easily create new cache types

Assigning quantification results

• Quantification results can be easily attached to input or output data
• Well-defined, fully-automated and portable way to assign quantification results to data for external tools
• Predefined data cache that acts as shared storage for quantified data

CMake build system integration

• Automated generation of installable shared libraries
• Automated generation of installable executables for worker modules

Support for quantified data

• Matrix data type including mathematical operations
• Values with unit
• • Unit conversions
• • Higher order units

Imaging capabilities

OpenCV integration

• Single-image cache
• Image stack cache
• • Allows access to each single-image cache
• • Easy parallelization

OME Files integration

• OME TIFF cache that allows access to individual OME TIFF planes
• • Can read OME TIFF format including metadata
• • Can write OME TIFF format including metadata
• Automated conversion from/to OpenCV images
• Read and write access to OME metadata such as microscopy parameters or physical size
• Support for OME quantity types
• • Including higher-order quantity types such as area and volume
• • Voxel/3D quantity types