AI for life science research

Automatic Image Processing
Smart Structure Recognition
AI and Machine Learning
Custom Protocols
Computer Vision

* Digitalize your in-house expertise


Automatic cells counting on z-slices with multiple markers

We offer very fast segmentation and classification of multiple, up to 10,…...

read more

Mixed-effects model

We offer statistical analyses for output data extracted from our image analysis. We…...

read more

Track cells on 2D/3D

Do you trace your cells on 2D/3D – z-stacks? We use a…...

read more

Sholl analysis

Automatic Sholl analysis of individual neurons. Segmentation and classification of each neuron…...

read more

Save the knowledge of your in-house experienced observers!

Use machine learning training to keep the knowledge in the team!...

read more

Mapping /stitching images with aligned segmentation!

Mapping is fully automated, but manual corrections are always possible if the…...

read more

KARMENstudio is Used by:


How do you analyze images in your research?

Labor-intensive? It’s painful?
Time-consuming? Inefficient?

  • we set custom automatic protocols for your set of data;
  • it’s fast, only sec/image
  • train the software with your experienced observer(s)
  • significantly increase your data harvest and analytics

Planning new project? With loads of images?

How/who will analyze it? Do not reduce your ambitions!
  • using the power of deep learning
  • set you a pipeline for FULL image analysis

Large database with images, -grab more from it?

It will take TOO LONG to go through the data again?
How/who will analyze it?
  • built custom tools for your automatic image analysis
  • follow up with your research needs,
  • significantly increase your data harvest and analytics


KARMENstudio is built for scientific & industrial DATA

The range and volume of data that can be imported and exported are significant.

We fully support the following file types:

  • Full-range images, full-range video formats;
  • 360° images from various camera types 
  • Document file formats including .doc, .pdf, .xls, .csv, etc.
  • 3D models (like .OBJ or .DAE, etc)
  • Hyperspectral layers/markers of the same fields (up to 100 layers)

We combine AI with other techniques to solve a problem

  • AQUACULTURE assets are complex, and the real world is noisy!
  • We spend time understanding the roots of the problem!

We combine multiple approaches and techniques to solve complex problems.

We create:

  • Physical models
  • Sets of image processing approaches
  • Multiple approaches and techniques such as:
    • Computer vision
    • Geo-Positioning
    • Our own Artificial Intelligence solutions

Private Label

  • You’ll never be boxed in.
  • We are all about customization!
  • Dedicated AWS server for each customer


  • delivering fully automatic inspection image/video analysis and reporting.
  • All solutions and deliverables are customized,
  • observations and data analysis are presented in various formats – oriented on need-to-basis.
  • In some use cases, KARMENstudio provides automated reporting. We generate the inspection report alone and only alert the human when there is an anomaly detected.

Training and testing phases

Training stage:

  • Personalized access to the KARMENstudio platform.
  • for the training of software using the power of machine learning!
  • keeps you in touch with data analysis - by 
    training the software by your experienced observer(s)
  • interactivity implemented which allows you fast performance
  • brings the efficiency of automatic image analysis even higher!
  • significantly increases your data harvest and analytics

Testing stage:

providing additional information regarding the expectations, probability, and efficiency of all the steps of analysis.

KARMENstudio workflow

The main workflow consists of base preprocessing, segmentation and classification with integrated functional modules and features, mapping and statistics.

  • The mapping module is merging individual images into one complete virtual image/folder.
  • Statistics module contains basic data analysis of main image features. Any extended data analysis is provided by additional suitable plugins.
  • Plugins supported: Image-based: include preprocessing step; Data-based: include data analysis; Based on image and data: include multiple image processing; Based on extended data: include structural analysis; Upgraded Functional Modules; Upgraded Features

Image analysis workflow includes following steps:

  • preprocessing,
  • segmentation,
  • classification,
  • multiple image processing,
  • structural analysis and
  • data analysis
  • Segmentation and classification contain intermediate computational scripts, each programmed as an independent functional module.

Deep Learning allows fast and efficient image analysis containing complex visual structures. In addition, thanks to Deep Learning, we can also quickly develop novel analysis scripts for new applications through automated proposing patterns of new scripts, based on our extensive experiential database, and adjusting the parameters of the new script modules.

Full analysis of a typical 8-megapixel image with resolution 3266×2450 pixels rarely takes more than 2 seconds on an average computer today.
Developing a novel script module for new applications takes only a few weeks.

Our image analysis is oriented to work with many related images, such as individual images forming a large visual scene, or a sequence of images captured in a short time, or slow time lapse of images, or images taken as cross-sections of 3D digitized objects.

This multiple-image orientation is ensured for high-speed image analysis and high data optimization, which enables simultaneous access and handling of even extremely large amounts of data.

KARMENstudio is highly modular and easily upgradable for different applications.
Upgrade models include plugins with new segmentation and classification functionalities, as well as plugin extensions for preprocessing, multiple image processing, structural analysis and data analysis.

Intuitive interface allows easy interactivity during all image analysis steps.
In the case of segmentation and classification it is possible to return to any previous intermediate step from the script, and to make adjustments to the parameters of the modules.

While the initial scope of KARMENstudio was developed for life-science microscopy, in particular in the field of cellular neuroscience, KARMENstudio has a much wider scope today, thanks to its distinct modularity and upgradability.

Examples of image processing modules used in KARMENstudio are: Structural analysis of bones and keypoint detection, Object tracing due to moving affinities, Point Spread Function detection, Validation by segment registration with other image analysis tools.

About us

"We are here to provide you AI ( machine learning ) tools for fast and objective image/video analysis!"

KARMENstudio team - committed to bringing solutions to help

We have a core team of 7 dedicated scientists, entrepreneurs, and developers. Led by:

Ana Bedalov

CEO Physicist

Saša Kijuk

Head Software Developer

  • 20 years of experience in image analysis
  • very skilled machine/deep learning developer

Tonći Kozina

Signal analysis engineer

  • 22 years experience in signal analysis, analytical operations, and development for military purposes.
  • rich experience in developing and maintaining software applications

Ana Bacelj


  • MSc astrophysics
  • very precize, selfdriven and creative image analysis specialist

Denis Zorić


  • 15 years of experience in logistic protocols and statistics
  • IP and transfer of technology in medical bionics
  • testing and verification of algorithms

Jelena Radić

Automatization engineer

  • with a fresh master in automatization engineering, she inherited all of our "problems" concerning full automatization of the processes and optimization of the usage of the server and human workers.  
  • Her pragmatic mind is conquering step by step our protocols and giving all of us a space to breathe!

KARMENstudio team -bringing custom smart structure recognition solutions!



You would have to:

– contact us:‬
-> Leave short instructions ( about  what do you want to measure from images/videos  )

– send us representative, but short image/video material

We will:

– build a custom protocol for your image/video analysis needs

– send you back DEMO materials :

  • with the short explanantion


You would have to :

  • Send us a short email and we will reply with a personalized link to our cloud to upload your image/video data.
  • Upload your first sample
  • Give us some instructions on at least one example image with marks and/or narrative explanation about what do you want to measure/gain from images.
    • We can also have a brief video call making sure we understand exactly what you would like us to “grab” from the images!

We will:

  • Build a custom protocol for your image/video analysis needs.
  • Send you back DEMO materials (images/clip, tables, stats,..):
  • We will be fast!

Contact us about pricing:

  • After the DEMO, we will send you a quote.
  • We charge building-up custom and trained automatic solutions.
  • Image/video analysis with ready custom solutions is later charged per image/video clip.
  • contact us!