The field of medical imaging encompasses a wide range of distinct technologies, and French startup Gleamer is expanding its focus to include magnetic resonance imaging (MRI) after previously developing AI-powered tools for X-rays and mammographies.

Rather than starting from scratch, Gleamer has strategically acquired two startups, Pixyl and Caerus Medical, that have existing expertise in AI-powered MRI analysis, allowing the company to accelerate its entry into this space.

Gleamer is part of a second generation of startups aiming to enhance medical imaging through artificial intelligence, following an initial wave of companies founded around 2014 and 2015. While many of these early startups did not succeed, there has been significant consolidation in the industry, with companies like Zebra Medical Vision and Arterys being acquired by Nanox and Tempus, respectively.

Founded in 2017, Gleamer has developed an AI-powered assistant for radiologists, which functions as a copilot for medical imaging. This tool enables radiologists to potentially improve diagnostic accuracy when interpreting medical images.

Gleamer’s software solution has been adopted by 2,000 institutions across 45 countries, with a total of 35 million examinations processed. The company has received CE and FDA certifications for its bone trauma interpretation product and offers CE-certified products for chest X-rays, orthopedic, and bone age measurements in Europe.

According to Gleamer co-founder and CEO Christian Allouche, a one-size-fits-all approach to radiology is ineffective. “It’s very complicated to have a large model that covers all medical imaging and delivers the level of performance expected by doctors,” he explained to TechCrunch.

To address this challenge, Gleamer has established small internal teams focused on specific areas, such as mammographies and CT scans. The company recently released its mammography product, which has been in development for 18 months and is based on a proprietary AI model trained on 1.5 million mammographies.

Gleamer has partnered with Jean Zay, the French government’s GPU cluster, and is working on CT scans for cancer detection.

However, when it comes to MRI, Allouche noted that it is a distinct technological space with various tasks, including detection, segmentation, characterization, and classification. To accelerate its entry into this area, Gleamer has acquired Pixyl and Caerus Medical.

The terms of the acquisitions have not been disclosed, but Allouche stated that the two companies will become Gleamer’s MRI platforms, with the goal of covering all use cases within the next two to three years.

Preventive medical imaging

While Gleamer’s models show promising results, they are not yet perfect. For instance, the company’s new mammography model can detect four out of five cancers, compared to a human radiologist without AI assistance, who typically identifies cancer in three out of five cases.

However, the productivity gains from a tool like Gleamer could revolutionize medical imaging. A missed tumor is likely to appear in a follow-up exam a few months later, making early detection crucial.

Allouche predicts that in the not-too-distant future, routine whole-body MRIs may become a standard practice, paid for by insurance companies, as they do not involve radiation.

However, the industry is already facing a shortage of radiologists in some cities, and a shift towards preventive imaging could exacerbate this issue. In such a scenario, AI tools like Gleamer’s could become essential for orchestrating and triaging medical imaging examinations.

Allouche believes that AI could automate the process of ruling out certain diagnoses, which is a significant portion of medical imaging examinations. “So, there’s a real need to automate all this with a very solid AI model that has a much higher level of sensitivity than a human,” he said.