Join 1363 founders getting motivational stories of how other founders started and grew their online businesses

Cerebriu: AI Solutions That Change the Face of Radiology


The health tech industry has been widely expanding for the past few decades.

The main culprit for the ongoing evolution of artificial intelligence? Expansion!

As health tech companies address the need for software and services in healthcare facilities, the use of AI for medical purposes has been highly emphasized. Even so, medical technology development is a lengthy and complex process, and departments like radiology have been falling behind.

Akshay Pai, Erik Dam, Mads Nielsen, Martin Lillholm, and Robert Lauritzen joined forces to develop an idea that improves efficiency and quality in the radiology field.

Supported by machine learning, Cerebriu uses a designated tool - Apollo Brain - to automate patient brain scans. It helps medical experts detect and prevent any signs of strokes, bleeding, or brain abnormalities. As a refined tool, Cerebriu enables maximum medical aid in diagnosing brain ailments.

Introducing AI within medicine is a bold challenge to take on - and here’s how 5 grand brainiacs pulled it off.

The Experts Team

Before the big five - Akshay, Erik, Mads, Martin, Robert - made it big in medicine, they were but mere beginners in business looking to create something extraordinary.

Akshay Pai obtained his Bachelor’s Degree in Electronics and Communication in 2008, from the Visvesvaraya Technological University. He further continued his studies at the University of Texas-El Paso and gained a Master’s Degree in 2010. Akshay is a reputable Doctor of Philosophy and Medical Image Analysis at the University of Copenhagen as well. He also worked as an Industrial Postdoctoral Researcher up until May 2021. To top it all off, Akshay is also the author and co-author of over 50 research publications.

Erik Dam, on the other hand, was born in Denmark. He obtained a Master’s Degree in Computer Science from the University of Copenhagen. In 2003, he also acquired a Ph.D. in Medical Image Analysis from the IT University of Copenhagen. Erik’s current role is that of an Associate Professor whilst he also serves as the Head of the Data Science Lab at the University of Copenhagen.

Mads Nielsen began his career as a Computer Science professor at the University of Copenhagen, a position he’s been holding for 14 years. He completed his Bachelor’s, M.Sc., and Ph.D. at the University of Copenhagen. In 1995, he acquired a Ph.D. in Computer Science as well. Mads is a true wiz in his field – with over 25 years of international academic experience in machine learning and medical imaging.

In 1998, Martin Lillholm just finished his M.Sc. in Computer Science, also from the University of Copenhagen. At the faculty’s IT sector, in 2004, Martin also obtained a Ph.D. Just like Mads and Erik, Martin also works as a professor at the Department of Computer Science at the same university.

Last but not least is Robert Lauritzen, a skilled IT science executive, working in the field for over 26 years. Aside from his IT background, Robert also boasts management experience obtained in high-tech medical startups and major pharma corporations.

Robert graduated from the Technical University of Denmark in 2004. Aside from his technical expertise, Robert is also a gifted polyglot with 4 languages under his belt - Danish, English, French, and German.

Transforming Radiology

Considering their vast experience and professional backgrounds, all five inventors dreamt to create something bigger than teaching. Crossing paths whilst teaching at university, it didn’t take the five plenty of time to develop a concept that would both pay focus to their main medical and IT skills, and serve the community as well. As Erik has been introducing the AI concept for radiologists for over a decade and had developed over 22 patents in his career, he became the unofficial leader of the project.

The fruits of their labor finally saw the light of day in 2018, when Cerebriu was founded.

Cerebriu is a Danish-based company that offers software solutions built on clinical research and deep learning and image qualification development. To do that, Cerebriu largely depended on conducting extensive research before the platform was launched.

They began their research journey in 2013 when the team engaged in detailed clinical monitoring. By 2014, the team had moved on to doing research on Hippocampal Texture - offering early detection of Alzheimer’s disease using MR hippocampal texture findings.

With a four-year gap, in 2018, two key research papers had been published: The first was on white matter lesions and the second, on brain segmentation. Following the platform’s launch, and with the arrival of 2019, the team carried out research for predicting and distinguishing various Alzheimer’s types and subtypes. That same year, they also carried out research on how to make the most of training data whilst using less of it.

Initially, the Celebriu team was faced with the challenge of insufficient training data used in machine learning. This marks one of the toughest hurdles not just for the platform but the industry as well. Namely, deep learning as a methodology is the provision of the artificial models’ raw data and training them to produce an output based on it.

Of course, the data assumed ought to be representative. But when it comes to limited data that is also available, its representation is put under question.

Another challenge for the Celebriu team occurred and it was an issue similar to the former. They struggled to develop a concept that would acquire just enough data to build representative models that work seamlessly. Needless to say, the obstacle is not only technical but financial as well. Let’s not mention that building data sets and acquiring them comes with a higher price tag.

Regardless of the challenges, the Celebriu team was ready to tackle every bump on the road. They developed a few innovative methods in 2019 to help cope with the issue and develop the kind of product medicine lacked.

The APOLLO Brain Tool

One of the approaches the Cerebriu team took in refining their AI model was to create a tool that tackles most of the work. APOLLO Brain is the result of technology and expertise, combined right. It automates brain MR-image acquisition protocols while informing both the patients and doctors if an acute stroke or bleeding is detected.

Furthermore, it also adapts the protocol of acquiring imaging targets for differential diagnosis. Apollo Brain is beneficial for everyone involved in the medical process – the hospital management, radiologists, clinicians, radiographers, and alas, the patients. The product use results in improved productivity in patients, faster decision-making, as well as reduced stress and anxiety.

Company Funding

Pulling off such expensive research and experiments, and acquiring the right equipment can only be achieved with sufficient funds and the right partners.

Cerebriu held 10 funding rounds and raised a total amount of $3.7 million. Their first funding round came from an Angel Round, held in July 2018, when only $110,000 was raised. Their most significant funding took place in 2020 via a four-investor seed round worth $960,000.

Cerebriu, Today

Cerebriu has its headquarters in Ballerup, Denmark, and counts a team of 30 board members and 13 employees. Cerebrui’s annual revenue is valued at $2 million. Facility-wise, the company services and patents - three total - are trusted by 14 hospitals, mainly based in Denmark.

Throughout its existence, Celebriu has formed partnerships with a few medical facilities in Norway as well, along with tech companies like Microsoft, Microsoft for Startups, Nvidia, and Hub.

In 2019, Celebriu’s team joined a Covid-19 research and aspired to create a new model. The model estimates the risk of newly diagnosed patients in need of intensive care and ventilator breathing support.

The company’s future plans include the continuation of its previous endeavors. They also remain focused on conducting more long-term research, especially in breast cancer and knee arthritis in patients.

Nothing spoke more of Cerebriu’s success than its official ISO-standard certification, granted in May 2020. Using deep-learning technology aided by over 50 years of relevant research, Cerebriu leaves no patient unattended, and ensures it puts the body’s computer - the brain - in superb health and performing to the nines!