Image cytometry still faces the problem of the quality of cell image analysis results. Degradations caused by cell preparation, optics, and electronics considerably affect most 2D and 3D cell image data acquired using optical microscopy. That is why image processing algorithms applied to these data typically offer imprecise and unreliable results. As the ground truth for given image data is not available in most experiments, the outputs of different image analysis methods can be neither verified nor compared to each other. Some papers solve this problem partially with estimates of ground truth by experts in the field (biologists or physicians). However, in many cases, such a ground truth estimate is very subjective and strongly varies between different experts. To overcome these difficulties, we have created a free web-based service called CytoPacq which is currently a flagship in our simulation-related services:
A web-based simulation framework capable of generating artificial microscopy image data fully in 3D+time. The simulated data imitate the image as if acquired using the fluorescence optical microscope.
Feel free to check this service and use the generated data!
CytoPacq can generate fully 3D digital phantoms of specific cellular components along with their corresponding images degraded by specific optics and electronics. The user can then apply image analysis methods to such simulated image data. The analysis results (mostly segmentation or tracking) can be compared with ground truth derived from input object digital phantoms. In this way, image analysis methods can be compared with each other and their quality (based on the difference from ground truth) can be computed.
If you need the synthetic image data for benchmarking immediately, please look into our repository of already pregenerated datasets which we call MUCIC (Masaryk University Cell Image Collection). You may find some of those data suitable for your needs.
We are currently running several concurrent projects focused on the synthesis of microscopy image data. Once, they are completed, they are gradually incorporated into CytoPacq as modules. Here, you can find the list of the projects:
A model of HL60 cell population that evolves in time. All the cells are modeled fully in 3D+time. They can move and split due to mitosis that is also simulated.
Included in CytoPacq: Yes
A model of a lung cancer cell with evolving filopodial protrusions. The cell is modeled fully in 3D+time.
Included in CytoPacq: Yes
This framework is capable of generating populations of fixed HL60 cell nuceli, granulocytes, or clusters of human color tissues.
Included in CytoPacq: Yes
A model of multiple, mutually interacting lung cancer cells. The cells are modeled fully in 3D+time.
Included in CytoPacq: Not yet
A model of large cell populations forming the tubular network. The model is limited to 2D space. Each cell is represented as a mask.
Included in CytoPacq: Not yet
A model of large cell populations forming the tubular network. The cell is composed of several subcellular structures (nucleus, mitochondiron, ...).
Included in CytoPacq: Not yet
A model of 2D image sequences showing living cell populations together with ground-truth images for the evaluation of cell tracking tasks.
Included in CytoPacq: Not yet
A deep learning-based model of HL60 cell shape. The cell is represented as a fully 3D volumetric binary mask.
Included in CytoPacq: Not yet
A deep learning framework for modeling of living cell shapes in 3D+time. Cell shapes defined by signed distance functions are implicitly represented by a neural network.
Included in CytoPacq: Not yet