Jul 15, 2026
A bad avocado can ruin a breakfast, according to a German start-up that claims to have industrialized MRI scanning for food production. Orbem, as reported by AgNavigator, says it uses artificial intelligence to speed up the analysis of scans, allowing it to inspect the interior of food items and significantly cut food waste across the industry.
Funding and expansion
In January, Orbem closed a Series B financing round of EUR55.5 million. The company plans to use the funds to expand into the United States and to move from poultry into fruit and vegetables. Orbem claims to be the first company to fully automate and industrialize MRI for commercial applications.
Technology and applications
Orbem is a deep-tech company founded in Munich that specializes in AI and imaging technology. The firm states that a typical MRI scan takes between 20 and 25 minutes, but it has reduced that time to less than one second and integrated automation equipment for industrial use. The technology is applied mostly to food, including eggs, watermelons, mangoes, avocados, and other fresh produce. For each type of food, Orbem says it can look inside in real time and determine whether the item is good or bad, assess quality, ripeness, maturity, breakage, or internal defects such as hollow heart in watermelons.
Orbem launched in poultry in January 2023 and reports that it has since scanned more than 200 million eggs. The company claims that this represents more MRI images than the global healthcare industry has produced. The data is used to refine its AI technology, and Orbem says it now holds the largest dataset of MRI images in the world.
Why food before healthcare
Orbem’s CEO, Pedro Gomez, explained to AgNavigator that the company started with food for two reasons. First, the impact on food waste and its implications for sustainability and greenhouse gas emissions is substantial. Second, the technological maturity required for Orbem’s approach was not yet sufficient for clinical applications. Gomez noted that he previously worked with humans, neuroscience, brain tumors, and multiple sclerosis, but the technology needed to evolve. In food, the requirements allow the company to validate the technology and demonstrate that MRI can run at high throughputs, enabling the collection of massive data that would be difficult to gather in a constrained clinical environment.
Reducing waste at production level
Gomez indicated that identifying a bad fruit at the retail level typically sends it to the trash. However, at the production level, different grades of fruit can still be used for juice or other by-products. He also pointed out that perfectly good fruit is often sent to processing simply because it looks imperfect on the outside. By showing what is inside, producers can market that fruit for its actual quality and recover margin they had previously written off.
Biological intelligence platform
Orbem’s technology goes beyond simple quality identification. Gomez said the company can detect the sex of an egg, which is an example of understanding biology. For fresh produce, biology informs quality, as the biochemical composition of an avocado manifests in how the flesh appears. By connecting biological data with other types of data, such as field or production data, Orbem aims to create more valuable intelligence. Gomez divided this into intelligence of the individual sample, which is about quality, and intelligence of the population, which relates to the biology of the wider group. This allows customers to understand why certain batches have different characteristics and to make better decisions.
Commercial model and scale
Orbem delivers the MRI equipment to places where it was not previously possible, such as hatcheries or production facilities. Its commercial agreements tend to be volume-based arrangements, through which it provides the intelligence customers need to inform their production processes. The company currently has 200 full-time employees and customers in eight countries. Gomez stated that revenue growth has exceeded 100% for each of the past couple of years since the launch in 2023. This year, Orbem launched with JimboFresh in Spain, one of the largest watermelon producers in Europe.
Competition and differentiation
Gomez acknowledged that Orbem competes with other sensors and measurement devices, such as hyperspectral sensors and optical systems. The key difference, he said, is that Orbem can see every detail inside an item. He noted that MRI has existed since the 1980s and that scanning biology with MRI is not a new idea. What is new, according to Gomez, is the ability to do it at a cost, speed, and scale that makes it industrially feasible, which has become possible only with the combination of AI and automation technology.
Long-term vision
Gomez described a long-term vision in which Orbem’s technology becomes the standard for food production, scanning hundreds of millions of kilograms of food per year. As the company builds the largest and most diverse imaging dataset in the world, it plans to apply the same foundational intelligence to human health. He noted that damage in a watermelon or avocado manifests in an MRI machine in the same way as damage to human tissue. The ultimate goal, Gomez said, is to become the intelligence company for food and health, bringing AI-powered MRI to both markets.