Mipar

Why choose Mipar?

Use Mipar to detect and measure features:

• Much faster and with much less variability in use than with manual analysis
• With greater accuracy than other automated solutions
• With built-in oversight of your results

Able to identify and measure the characteristics of all types and formats of images:

• Microscopes
• Scanners and radiographers
• Satellites
• Cameras
• Smartphones
• Drones

For students and universities

MIPAR is used by universities around the world to train students in advanced image analysis tools for materials science, life science, etc. Students have access to affordable prices on a monthly or semester basis. Departmental and university licenses are also available. If you are interested, we invite you to send us a quote request.

The MIPAR Academy is a comprehensive self-study course designed to get users up and running quickly while building lasting skills and expertise.

Post your questions and comments on the MIPAR User Forum to get answers from community members and MIPAR application experts.

Student Use Condition:

For use only by students on student-owned systems for academic courses and research at degree-granting institutions.

This offering is not for commercial, government or other organizational use.

If you wish to use MIPAR in a business, government lab, as an instructor at a university, for research, or for commercial or industrial purposes, please request a quote.

License option*: fixed seat, specific computer.

*The computer must be owned by the student, not the university.

Domains of application

Life sciences 

Fluorescence

Mitochondrial distribution in a differentiated PC12 cell: Fully automatic detection of mitochondria, separation of neurites and cell body, and measurement of mitochondrial fraction in each.

Histology

H&E staining of odontogenic keratocystic tumor epithelium: Quantification of epithelial layer thickness, after fully automated layer identification.

Petri dishes

Bacterial colonies on a petri dish, automatic plate detection, estimation of the number of colonies that have grown together.

Materials sciences

Silica nanoparticle size and shape

Fully automated solution for nanoparticle size and shape characterization. Quantify a set of samples from a single scan or batch process.

Analysis of porosity and thickness of nanofibers

Quantify fiber density, void porosity and fiber thickness distribution.

Acrack analysis, part defect identification

Quantify additive manufacturing crack density, size distribution, localized density. Fully automated solution for single field or whole surface point quantification.

Inspection and control

Automated inspection of railroads using drone imagery

Track components quantified from drone images as part of a weekly railroad inspection mandated by the Federal Railroad Administration (FRA).

Aerial study of the severity of deforestation

Percentage of forested (and deforested) areas automatically quantified from analysis of imagery produced by drones.

Contaminant control

Classification and size measurement of filter paper contaminants: The detected contaminants are classified into predefined types and the size and shape are reported.

Features

Spotlight (New)

Directly integrated into MIPAR, Spotlight is a cutting-edge tool for automating image analysis using AI. This turnkey solution works without designing or training an analysis model: simply load your scientific image to initiate the analysis. 

• Ease of use: unlike Deep Learning, no need to train a model. Spotlight is ready to use.
• Existing database: Spotlight relies on a preconfigured database that automatically detects objects, from the simplest to the most complex.
• Speed and Accuracy:  Get fast results with unparalleled accuracy of any desired characteristics.

Snap, powered by Spotlight

Snap is a smart drawing tool available in  Spotlight that automatically selects elements and features of interest on an image.

No need to make freehand annotations, just use the Snap box to automate the selection. 

Find out more about Spotlight: replay webinar Spotlight

Checkpoint (New)

Fully integrated into the MiPAR ecosystem, Checkpoint is an advanced tool designed to ensure compliance with 21 CFR Part 11 and GMP Annex 11. Tailored for demanding industries, Checkpoint guarantees the traceability, security and integrity of your image analysis data.

Easy to use and customizable to your specific needs, Checkpoint streamlines automatic detection of characteristics, measurement and reporting of results while ensuring complete traceability and validation by electronic signature.

• Regulatory compliance: Checkpoint ensures compliance with 21 CFR Part 11 and GMP Annex 11. You can use integrated tools for detection, measurement and reporting to meet regulatory standards.

• User access control: You can manage user access with connection control and group management, track project history, and control result exportation for maximum security.

• Guaranteed traceability: Every step of the workflow is recorded and accessible for full auditability and total compliance.

Deep Learning

• Trace: Trace your features on some images. Even use the plots you already have.
• Train: A deep neural network learns to recognize what you have plotted.
• Run: Run your model on a new image to detect complex features.

The features of Deep Learning: 

• Adjustment of certain characters (e.g.: specific delimitation).
• 5 to 10 image analyses are enough for learning and recognition.
• From pre-established parametric forms followed by a recording 

Example usage

Detecting kernels while ignoring twins has been a challenge for the community for decades. Mipar’s Deep Learning function has successfully met this challenge.

Graphic display

• Visual exploration of live image quantification. 
• Exporting image data, graphs or digital tables ( individual or report)

 

Batch processing

• Ability to batch analyze quickly and accurately.
• Ability to make corrections if necessary.
• Graphic visualization in the form of a histogram of the analyzed batch.

Here’s how to set up and run a batch process to segment and measure features from a set of images.

Required configuration

The MIPAR API allows executing recipes in Python scripts and applications. Access to the API is granted to all perpetual licenses at no cost. Working in conjunction with MIPAR image analysis software, users can prototype, refine and deploy scalable image analysis solutions. The library can be deployed both locally and over a network.

Network deployment

The MIPAR API can be deployed on a network of servers as part of other networked applications, such as a networked database. Images stored in a database can be automatically quantified and archived.

How does this work?

• Upload the recipe you want.
• Open your image in the image processor.
• Pull and drop your recipe.