CryoVision is an AI-powered software platform for automated analysis of microscopic cell images. Formerly known as the CellsCalculator project, it has been refactored and improved to provide even better performance and user experience.
We eliminate one of the biggest bottlenecks in modern biotech:
slow and error-prone manual cell analysis
With CryoVision, researchers can process complex 2D and 3D cell cultures in seconds instead of hours, unlocking faster discoveries in medicine, drug development, and bioengineering.
- ⏳ Up to 10 hours to analyze a single image manually
- 🎯 Only ~80% human accuracy
- 🧠 Scientists wasting time on repetitive work instead of research
- 🧬 3D cell structures are extremely difficult to analyze manually
CryoVision is a ready-to-use AI application that:
- 🔍 Automatically segments cells and tracks spheroids
- 📊 Extracts morphology (area, diameter, volume)
- 🧠 Builds 3D spheroid reconstructions
- ⚡ Processes images in under 1 second
- 💻 Runs on entry-level laptops (no GPU required)
- 🚀 Up to 7,000× faster than manual analysis
- 🎯 85.2% accuracy with consistent results
- 🔁 Fully reproducible (no human bias)
- 🔧 Works across different cell lines & setups
- 📈 Scales from small experiments to high-throughput screening
CryoVision combines modern AI and computer vision techniques:
- 🧠 Deep-learning-based instance segmentation
- 🔗 Custom object tracking algorithm
- 📐 Morphological feature extraction
- 🧩 Modular architecture for scalability
All wrapped in a simple, user-friendly interface — no coding required.
Should you want to try the CryoVision application, follow the guidelines below:
- On project's GitHub main page, search for Releases section in the right menu;
![[Releases]](/EugenTheMachine/CryoVision/raw/main/images/Screenshot_8.png)
- Click on the latest release available;
- In the opened release window, go to Assets section at the bottom and click on
CellsCalculator.zipfile archive;![[Assets]](/EugenTheMachine/CryoVision/raw/main/images/Screenshot_9.png)
- Wait until the archive is completely loaded and then unpack it;
- Download the models following the instructions and place them into the
modelsdirectory; - Enter the automatically created
CellsCalculatorfolder and run themain.exefile. - Enjoy the application!
![[Releases]](/EugenTheMachine/CryoVision/blob/main/images/Screenshot_8.png)
![[Assets]](/EugenTheMachine/CryoVision/blob/main/images/Screenshot_9.png)
Remember: it is forbidden to rename or change location of ANY elements within the CellsCalculator directory, as it may badly influence the application behaviour, up to its full crash with unpredictable errors on the way.
Should you want to run the raw application code, follow the guidelines below. Note that the prompts are designed for Windows CMD - for bash you may need to use other syntax:
- Clone the repository using the prompt below:
git clone https://github.com/EugenTheMachine/CryoVision.git- Enter the folder in which the repo is cloned and set up your environment. To do that, you need to have Python 3.7 or above pre-installed. Install the required dependency packages by running:
pip install -r requirements.txt- To start the application, execute the
main.pyfile - you can do that by running:
python main.py- Enjoy the application running!
CryoVision has emerged as a student project and was developed by the following team:
- Ye. Ponomarov - Consultant, former Team-Lead, former ML-Engineer;
- S. Lytvynenko - Team-Lead, ML-Engineer;
- O. Kovalov - Tech-Lead;
- K. Noskova - ML-Engineer;
- R. Malakhov - ML-Engineer;
Our team also deeply appreciates continuous support from our mentors whose consultancies have made our journey much easier:
- M. Tatariants - Senior Lecturer at the Department of Computer Mathematics and Data Analysis, NTU "KhPI"
- V. Kolbasin - Senior Lecturer at the Department of Computer Mathematics and Data Analysis, NTU "KhPI"
- G. Bozhok - Senior Researcher at the Institute for Problems of Cryobiology and Cryomedicine of the NASU
Feel free to reach out to us at Oleksii.Haluza@khpi.edu.ua