Gleamer Expands into MRI with Strategic Acquisitions | Radiology AI

Gleamer Expands into MRI Analysis with Strategic Acquisitions

Medical imaging represents a diverse field, incorporating numerous distinct technologies. Having previously developed AI-driven enhancements for X-rays and mammograms, the French company Gleamer is now directing its focus towards magnetic resonance imaging (MRI).

Acquisition and Merger Strategy

Rather than initiating development independently, Gleamer has opted to acquire Caerus Medical, a startup already engaged in AI-powered MRI analysis, and is simultaneously merging with Pixyl. This approach allows for accelerated progress in the MRI domain.

A Second Wave of AI in Medical Imaging

Gleamer is positioned within a second generation of startups aiming to refine medical imaging through the application of artificial intelligence. Initial ventures in this area emerged around 2014 and 2015. While many did not succeed, the industry has witnessed consolidation, exemplified by the acquisitions of Zebra Medical Vision and Arterys by Nanox and Tempus, respectively.

Gleamer's AI Assistant for Radiologists

Established in 2017, Gleamer has been developing an AI assistant designed to support radiologists, functioning as a collaborative tool for interpreting medical images. This technology aims to potentially enhance diagnostic accuracy.

Current Adoption and Certifications

The startup’s software solution is currently utilized by over 2,000 institutions spanning 45 countries. To date, Gleamer has processed a total of 35 million examinations. The company has secured both CE and FDA certifications for its product focused on bone trauma interpretation.

Furthermore, within Europe, Gleamer offers products tailored for chest X-rays, orthopedic assessments, and bone age measurements, all bearing CE certification.

The Limitations of a Universal Approach

“A standardized approach to radiology proves ineffective,” explains Christian Allouche, co-founder and CEO of Gleamer, in a statement to TechCrunch. “Developing a single, comprehensive model capable of addressing all medical imaging needs and meeting the performance expectations of physicians is exceedingly complex.”

Specialized AI Teams

Consequently, Gleamer has established dedicated internal teams concentrating on specific modalities like mammographies and CT scans. “Our mammography product, the result of 18 months of work, was released three weeks ago,” Allouche stated. This product leverages a unique AI model trained on a dataset of 1.5 million mammographies.

Leveraging Government Resources

“We benefit from a partnership with Jean Zay, the French government’s high-performance computing cluster,” Allouche added. The company is also actively developing CT scan applications for cancer detection.

The Complexity of MRI

However, the challenges presented by MRI are distinct. “MRI represents a different technological landscape,” Allouche clarified. “It encompasses a wide range of tasks, extending beyond simple detection to include segmentation, characterization, classification, and the analysis of multi-sequence imaging.”

Accelerating MRI Development

To expedite progress in the MRI field, Gleamer is acquiring Caerus Medical and merging with Pixyl, both companies with established expertise in this area. The financial details of these transactions remain undisclosed.

Future Ambitions

“These two companies will serve as our foundational MRI platforms, with the objective of encompassing all relevant use cases within the next two to three years,” Allouche concluded.

Proactive Medical Imaging Techniques

Although Gleamer’s AI models demonstrate considerable potential, they haven't yet reached complete accuracy. Specifically, their recently developed mammography model is reported to identify cancer in four out of every five instances. This contrasts with human radiologists, who typically detect cancer in three out of five cases without the aid of artificial intelligence.

The efficiency improvements offered by a system such as Gleamer have the capacity to significantly transform the field of medical imaging. A tumor that is initially overlooked is likely to be identified during a subsequent examination conducted a few months later.

Allouche predicts a future where regular, full-body MRI scans become commonplace, covered by insurance plans due to their non-irradiating nature. “I believe that in the foreseeable future, we will all undergo routine whole-body MRIs, with the cost being borne by our insurance providers,” he stated.

Currently, certain urban areas are experiencing a shortage of radiologists, struggling to keep pace with the existing demand for reactive imaging services. A transition towards proactive imaging will necessitate the widespread adoption of AI-powered tools.

The CEO of Gleamer envisions AI functioning as a central “orchestrating and triaging” system. A significant proportion of medical imaging tests are performed to eliminate potential diagnoses. “Therefore, there is a substantial requirement to automate these processes using a robust AI model exhibiting a greater degree of sensitivity than a human expert,” Allouche explained.

The Role of AI in Medical Imaging

• AI can enhance cancer detection rates.
• AI can improve the efficiency of radiologists.
• AI can facilitate the shift towards preventive care.

The increasing demand for imaging services, coupled with radiologist shortages, highlights the necessity of AI integration. A highly sensitive AI model is crucial for automating the initial screening process.