Unmet Need
Biological research relies on accurate and precise image analysis, from tracking cells over time, assessing changes to organelles under experimental conditions, to quantifying localization of immunofluorescence. As imaging technology advances and biological questions become more detailed, a massive amount of high-resolution data is acquired that requires segmentation, labeling, annotation, and analysis. However, most existing processes to do so are time-intensive, cannot handle large volumes, or cannot handle nuanced segmentation (Taha et al.). Therefore, there is a need for improved automated image analysis with high resolution segmentation and labeling to accelerate accurate biological research and progress.
Value Proposition
· Reliably scalable, even with large structures, and multi-channel, time-lapse, or 3D images
· Capable of multiple methods of cell and structure tracking with automated alignment
· Faster and easier to utilize post-analysis data
· Can be combined with user-defined functions for more detailed analysis
· Capable of multi-object tracking with respect to other objects within a standardized single cell coordinate system
Technology Description
Researchers at Johns Hopkins have developed an automated backend solution for single-cell image analysis capable of single-organelle segmentation. Utilizing machine learning models or traditional image processing techniques, the package is capable of cell identification and isolation, as well as structural labeling and the application of detailed analysis as specified by the end-user. This solution is scalable and allows for faster analysis of large datasets over time or in three dimensions.
Stage of Development
The software package is fully functional as a command-line package. Looking for partners to integrate the package into an easier-to-use user interface.
Data Availability
Demo available at mitodynamics.net
Publication
N/A
