Domain Registration

Israeli startup Ibex Medical aims to transform cancer diagnostics with AI tech

  • October 19, 2021

Advanced technologies such as artificial intelligence (AI) and machine learning have been reshaping the healthcare landscape, introducing new processes in fields like medical imaging, disease screening and diagnosis, drug discovery and development, therapeutics, and medical research.

In the field of pathology, Israeli medical tech startup Ibex Medical Analytics has been drawing an enormous amount of attention for its AI-based cancer diagnostic software that helps pathologists better detect and grade different cancers.

Receiving an accurate diagnosis by a pathologist is vital to setting a treatment path for patients. Misdiagnoses or misgrading can often lead to unnecessary delays in treatment, sometimes with devastating consequences.

Ibex developed a software platform called Galen that uses AI-based algorithms to help pathologists analyze biopsies and improve the quality of cancer diagnosis. The company says the tech implements real-time quality control, and can help reduce diagnosis time, assist pathologists in getting a handle on workflow and even boost productivity. The software can be used to detect and grade prostate cancer, breast cancer, and, most recently, gastrointestinal cancer, with plans to expand the applications more widely.

The Galen platform offers the Ibex First Read, an application that analyzes cases prior to human pathologist review, enabling case prioritization, and the Ibex Second Read, a system that works in parallel with human pathologist review to identify any discrepancies.

In the five years since Ibex was first established by entrepreneurs Joseph Mossel and Chaim Linhart, its technology has been integrated into everyday clinical practice at major pathology institutes in Israel, the UK, and the EU, and has nabbed a breakthrough device designation by the US Food and Drug Administration (FDA) in June, as well as a CE mark for the European market.

The FDA designation will help Ibex fast-track the clinical review and regulatory approval of its technology in the US, where the company is now focusing its efforts, with an office in Boston headed by Mossel.

At its core, the Galen platform mimics the work of a pathologist with the benefits of AI tech for added accuracy and efficiency, Mossel and Linhart told The Times of Israel in a recent interview from Boston and Tel Aviv, respectively.

Basic pathology is “looking into the microscope at biopsies, a practice that’s been around for 100 years,” explained Mossel, who serves as Ibex Medical’s CEO.

The field has gone through two significant revolutions in recent years: the digitization of pathology labs, and the introduction of AI and deep learning which has started to flourish and is now accelerating, Linhart said.

“When we started five years ago, we weren’t sure if we could do it [bring this solution to life],” added Linhart, but the technology is now helping to save lives.

In the first week when Ibex started working with the Pathology Institute of Maccabi Health Services, Israel’s second-largest HMO, in 2018, the software “accurately identified a case that was missed by a pathologist, and this now happens every week,” said Mossel.

“Pathologists are human and they can miss or misgrade cancer, it happens,” he said. With the Galen platform, the system can identify more cancers, especially more complex cases, and help with a workload that is only increasing, he added.

Clinical studies in recent years have shown that the Galen platform can produce highly accurate results. The accuracy of Galen’s prostate cancer detection was found to be the highest, so far with the sensitivity measured at 98.46% and specificity at 97.33%, in a study conducted at the University of Pittsburgh Medical Center (UPMC) and reported by The Lancet Digital Health, a top-tier peer-reviewed clinical journal.

A separate study conducted with Medipath, the largest network of pathology labs in France, that comprised 100 patients who underwent prostate biopsy and were diagnosed as not having cancer, showed that the Galen platform identified 12 patients who were misdiagnosed.

“There are some things that humans just don’t see, and when you combine these tools, you get a more accurate picture,” said Lindhart

Mossel said Ibex’s vision is for the Galen platform to become “an indispensable tool for pathologists” where the software will become part of the workflow.

This week, Ibex announced a major partnership with Swiss multinational healthcare company Roche to jointly develop an embedded image analysis workflow for pathologists where they can access Ibex’s AI algorithms, insights, and decision-support tools using an internal Roche software called NAVIFY Digital Pathology. NAVIFY is the cloud version of the multinational’s uPath enterprise software developed for pathology labs, to introduce connectivity tools and automation.

“With the addition of Ibex’s clinical-grade image analysis tools to our NAVIFY Digital Pathology menu, we can aid pathologists and providers in delivering value-based patient care by increasing their efficiency and accuracy for higher-quality cancer diagnosis,” said Jill German, head of Roche Diagnostics Pathology Customer Area, in a statement on Monday.

Mossel said that, together with Roche, Ibex was looking forward to “transforming pathology by delivering powerful AI-based tools to the fingertips of pathologists.”

Mossel and Linhart said Ibex sells the Galen platform to pathology institutes, large hospitals, and academic and private labs all over the world and that the feedback so far has been excellent.

Currently, the startup is focused on expanding its reach in the US.

Ibex employs about 40 people across its offices in Tel Aviv, its tech base, and Europe, where its commercial team is based. The startup has raised over $50 million with investors such as Dell Technologies Capital, the corporate venture arm of Dell Technologies, Israeli medtech fund aMoon, Octopus Ventures, Planven Entrepreneur Ventures, and 83North.

Article source:

Related News