Ihr Browser (Internet Explorer 10/11) ist veraltet. Aktualisieren Sie Ihren Browser für mehr Sicherheit, Geschwindigkeit und den besten Komfort auf dieser Seite.
Minimic microscope

Minimic – Intelligence for routine microscopy

For over 100 years, microscopy has been one of the most important examination tools in medicine. Artificial intelligence now enables microscopes to learn from human experts and make diagnoses independently. At Intuity, we want to rethink the role of microscopes, turning the device itself into a medical expert.

Screening Thousands of Cells to Diagnose Malaria

Back straight, ticker in one hand, the focus wheel in the other—Jessica sits in front of the microscope, staring through the eyepieces. Thanks to the 100x magnification, she can see the single cells in a drop of blood that she had stained and put onto a glass slide. Some of the cells carry small, violet speckles. These are malaria parasites that reside in red blood cells, feasting on the hemoglobin. Jessica spends several hours at the microscope, scanning dozens of samples, counting infected red blood cells. Every day.

Jessica works in the research group of Prof. Dr. Friedrich Frischknecht at Heidelberg University Hospital. The research of the group focuses on understanding the lifecycle and proliferation of the parasite causing malaria. For the experiments carried out in the lab, it is crucial to determine the stage of infection and the percentage of infected blood cells in the samples. Consequently, counting cells under the microscope is a time-consuming part of the everyday routine for Jessica, the lab technician, as well as all master and PhD students in the lab. “This is not just done in my lab, but it’s done in hundreds of labs around the world as well as in hospitals to diagnose the parasite. Malaria microscopy takes an unbelievable amount of time and of course slows down scientific progress and critical care”, Dr. Frischknecht explains.

„More than 204 million patients were tested for malaria by microscopic examination in 2016.”
WHO Malaria Report 2016

Artificial Intelligence as Enabling Technology

Why shouldn’t a microscope do the counting and analysis all by itself? With this unconventional approach, Intuity started to address the problem of manual routine microscopy in 2017. The hypothesis: increasingly reliable artificial intelligence (AI) algorithms together with compact and powerful sensors and microprocessors provide a comprehensive set of tools for the development of affordable yet intelligent laboratory instruments.

The result of the project after two years work: the Minimic. “Our Minimic is a compact, comparatively inexpensive light microscope. It’s intelligent because it does not only magnify tiny objects, it also detects and counts them,” Markus Turber, designer and founder of Intuity Media Lab, explains. The Minimic is a seamless solution from the microscopic sample to the analysis result. Thanks to the integrated AI-powered image analysis module, the minimic helps saving time and ensures high-quality results independent of the user’s experience in microscopy.

Counting cells at the microscope is part of everyday life in research and medicine.
Counting cells at the microscope is part of everyday life in research and medicine.
The Minimic is an intelligent laboratory device for the Workbench.
The Minimic is an intelligent laboratory device for the Workbench.
The first small series grows up the Intuity workshop.
The first small series grows up the Intuity workshop.
The first tests with tuberculosis samples on the Minimic.
The first tests with tuberculosis samples on the Minimic.

Simple Hardware, Intelligent Software

Even at first glance, it is evident that the Minimic breaks with the standards of classical microscopy. There are no eyepieces for direct observation of the sample. Compared to the heavy cast iron look of most microscope bodies, the Minimic, which is barely the size of a shoe box, appears surprisingly dainty. The reason for this: hardware and optics were tailored for digital imaging instead of visual inspection. Further, instead of using elaborate and expensive technology to achieve high absolute precision, the team at Intuity relied on inexpensive components combined with intelligent software control.

Livelong Learning—Also Applies to the Minimic

The potential impact of intelligent microscopy systems is huge since malaria is by far not the only disease diagnosed through manual microscopy. A large number of infections, including tuberculosis, are microscopically identified in blood and other body fluids. In order to explore the potential of the Minimic and train it for the first applications, Intuity just finished assembling a mini-series of ten microscopes in the company’s own workshop. These are now being tested together with laboratories and research groups, among others, of course, with Dr. Frischknecht in Heidelberg.

The trainability and flexibility of the Minimic-AI enables easy adaptation of the system to other fields of application beyond medicine and life sciences. “In the interaction with pilot users we realized how versatile our approach to microscopy actually is: We started with the diagnosis of malaria in blood samples, now we are talking about parasites in veterinary medicine, pollen content in honey, and dust particle distribution on industrial products. We are curious where this journey will take the Minimic next”, said Mr. Turber.

Next Steps…

Intuity currently focuses on exploring new use cases for the Minimic and is looking for investors and partners to bring the Minimic to the market.

If you would like to learn more about the Minimic project and its further journey, please subscribe to the Minimic newsletter at minimic.com.