You would like to learn more about our products or have questions? Contact us, we will be happy to help you!

Call us+49 89 89 80 550

Write to usinfo@mail.de

Find the right imaging module with our online ProductFinder.

Search now

AI Microscope

AI microscope for urine sediment examination

with CNN-based image processing system from Basler and Opto

In a urine sediment analysis, laboratory staff examine the solid components of a urine sample under a microscope to diagnose diseases of the kidney or urinary organs. However, manual examinations under the microscope are usually error-prone and take up valuable time. Many laboratories therefore use automated systems. These differ in terms of the required performance, functionality, complexity and size, and thus also in terms of the integrated image processing system. The AI digital microscope from Opto shown below is equipped with a Convolutional Neural Network (CNN)-based embedded vision system from Basler. The low-cost, high-performance microscope is an important first step towards automating laboratory analyses and processes.

What is the problem?

Laboratory analyses must increasingly be provided quickly and cost-effectively, naturally without compromising reliability. The decisive factor is the integration of the right, i.e. coordinated, hardware and software components. Here, it is realized with a CNN-based analysis algorithm on the appropriate embedded vision system from Basler, completely integrated into a compact, robust Profil M imaging module from Opto.

The solution

The demo provides a reliable and cost-effective CNN-based imaging system suitable for tasks such as automated microscopic examination of urine sediments. 

The uniqueness of the solution lies in the integration of a high-performance transmitted light microscope from the Opto-Imaging Module family and the extremely powerful vision system consisting of a 5 MP color camera module and the embedded vision processing board - both developed by Basler. The core element of the price/performance optimized board is the i.MX 8M Plus from NXP®. 

The digital microscope features 20x magnification (200x for a standard desktop instrument) and an integrated transmitted light LED condenser optimized for urine sediment analysis applications. The camera and processing board are fully integrated into the microscope's aluminum frame and can be controlled via a single-cable USB 3.0 connection. The result is an ultra-compact, edge-processing AI microscope sensor for use in high-throughput machines or in a point-of-care (POC) analyzer.

The AI microscope presented here will be equipped with an intuitive user interface for urine sediment analysis and a pre-trained CNN for key markers to classify the urine sample for further diagnostics. The on-board software coordinates image acquisition and processing and controls the LED. The camera module used has an image signal processor (ISP) that performs all steps of image preprocessing. The subsequent inference of the CNN is performed on a special processor of the i.MX 8M Plus System on Chip (SoC), the so-called Neural Processing Unit (NPU). Using hardware acceleration, the NPU handles the typically computationally intensive tasks for the CNN, making execution particularly efficient. 

To make this package plug-and-play ready, Opto GmbH offers customized training assistance for the CNN and support for integration into local network architectures as a service for OEM integration. We can adapt existing algorithms to automate specific pre-classification of patient samples. 

Your benefit

 
  • Compressed microscope and vision know-how from specialists Basler and Opto in a plug&play AI microscope.
  • Expertly manufactured with long-term availability and price/performance optimization 
  • Seamless integration of traditional and CNN-based image processing and analysis 
  • One-stop solution for automating your individual microscopic process