Medical imaging is a wide term that includes several separate technologies. After working on AI operating tools to strengthen X -rays and mammography, French start Ablaze It is now aimed at tackling magnetic resonance imaging (MRI).
Instead of starting from scratch, Gleamer has acquired two newly established companies that have already worked in MRI resolution that is powered by AI: Pelvic and Medical.
Gleamer is part of the second wave of newly established businesses trying to improve medical imaging using artificial intelligence. Many technology founders created newly established businesses around this issue in 2014 or 2015. While most of them did not go anywhere, there was some integration in the field. For example, both Arterys Zebra were acquired nanox and Stormrespectively.
Founded in 2017, Gleamer has built an AI assistant for radiologists, a kind of copilot for medical imaging. With Gleamer, radiologists can theoretically improve diagnostic accuracy in the interpretation of medical images.
The start has already convinced 2,000 institutions in 45 countries to use the software solution. Overall, Gleamer has edited 35 million exams. The company has received CE and FDA certifications for the bone interpretation product. In Europe, it also offers products that focus specifically on X -ray measurements, orthopedic and CE certification.
“Unfortunately, the approach of a size that fits all radiology does not work,” said Gleamer Christian Allouche’s co -founder and chief executive in TechCrunch. “It is very complicated to have a large model that covers all medical imaging and offers the level of performance expected by doctors.”
That is why the company created small internal groups focused on mammograms and CT scans. “Three weeks ago our mammography product was released, which we have been working on for 18 months,” Allouche said. It is based on a privately owned AI model trained in 1.5 million mammograms.
“We have a collaboration with Jean Zay, the French government’s GPU complex,” Allouche said. The company also works in CT cancer scans.
But what about magnetic resonance imaging? “Magnetic resonance imaging is a different technological space,” Allouche said. “You have many tasks in magnetic resonance imaging. It’s not just detection. You have appreciation. You have a detection, you have a characterization, sorting, multiple sequence visualization.”
That is why Gleamer buys two small businesses working in this area for several years to move faster. Gleamer does not reveal the terms of the agreements.
“These two companies will become the two MRI platforms, with the clear ambition covering all cases of use in the next two to three years,” Allouche said.
Preventive medical illustration
While Gleamer models show many promising results, they are not perfect yet. For example, with the company’s new model of mammography, startup claims that it can detect four in five cancers. In comparison, a human radiologist without AI usually identifies cancer in three of the five cases.
However, productivity profits from a tool like Gleamer could radically change medical imaging. A lost volume is likely to appear in a follow -up exam a few months later.
“In the non-distant future, I think we will all get routine the whole MRIS body paid by our insurance companies-since it is not radiating,” Allouche said.
However, in some cities, there are already very few radiologists to meet the demand for reactive visualization. If the industry shifts to preventive imaging, AI tools will become necessary.
Gleamer’s CEO believes AI could become a “orchestration and classification” tool. Most medical examinations are performed as a way to exclude certain diagnoses. “So there is a real need to automate all this with a very compact AI model that has a much higher level of sensitivity than a human being,” Allouche said.
