Tel Aviv-based startup MedyMatch Technology has reached a five-year licensing agreement with IBM Watson Health that will enable the artificial intelligence heath arm of the US giant to market the Israeli technology together with its other services.

MedyMatch, which aims to introduce cognitive tools to the daily workflow of emergency rooms, has developed an artificial intelligence (AI) software that aims to be more accurate than the human eye and help physicians more quickly assess patients suspected of head trauma or stroke, ruling out the presence of a bleed in the brain.

IBM Watson Health uses AI to help doctors and hospitals better diagnose illnesses and cure diseases. The company now plans to integrate the MedyMatch product in the services it offers hospitals globally, using the Watson brand. The software will be offered to experts working in hospital emergency rooms and other acute care settings to help them identify cases of intracranial bleeding as a result of head trauma and stroke.

The original equipment manufacturer (OEM) deal is expected to generate several million dollars in revenues for MedyMatch over the coming five years, said Michael Rosenberg, chief commercial and financial officer of the company, in a phone interview, and the market opportunity ahead is tens of millions of dollars annually for the coming five years.

MedyMatch’s first product is expected to be commercially available at the end of this year, Rosenberg said.

Picture of Israel's MedyMatch stroke assessment. (Courtesy)

Picture of Israel’s MedyMatch stroke assessment. (Courtesy)

The MedyMatch algorithm uses sophisticated deep learning, machine vision, patient data, and clinical insights to automatically highlight for a physician regions of interest that could indicate the potential presence of cerebral bleeds. In these cases, for every minute that passes more brain cells die, so a quick diagnosis is essential.

With deep learning, a series of examples are fed to the computer to set benchmarks for what is considered a baseline reading, Gene Saragnese, a former CEO of Philips Imaging Systems who joined MedyMatch as CEO in February 2016, said in an interview last year. Then a series of images is uploaded to the computer – and the machine learns what a bleed looks like from those images. Basically, “you are training the computer with examples in a way that it can then start to read images by itself,” he said.

MedyMatch's CEO Gene Saragnese. (Courtesy)

MedyMatch’s CEO Gene Saragnese. (Courtesy)

MedyMatch has secured billions of images from millions of cases via collaborations with hospitals in Israel and the US, including Hadassah Medical Center in Jerusalem and Massachusetts General Hospital in Boston. Those experts are helping train the software to read, Saragnese said.

According to the American Heart Association and American Stroke Association (AHA/ASA), stroke is the fourth leading cause of death and one of the top causes of preventable disability in the United States. Affecting 4% of US adults, it is forecast that by 2030, there will be approximately 3.4 million stroke victims annually in the US, costing the healthcare system $240 billion a year.

“At MedyMatch, we believe improvements in the interpretation of data will lead to better decisions, better decisions will lead to better outcomes for patients and lower cost for healthcare,” said Saragnese in a statement released Thursday. The collaboration with IBM will bring artificial intelligence to the bedside of patients, he said, and that is where “the future of healthcare lies.”

More deals to come

The IBM deal, the first licensing agreement for MedyMatch, is not exclusive, said Rosenberg, and the Israeli startup is in talks with other potential partners as well. Additional deals are expected shortly, he said. There are no talks at the moment of IBM acquiring the company, Rosenberg said.

“The implementation of AI-based computer aided detection and clinical decision support tools to medicine in general, and to the emergency department, in particular, has the potential to increase the speed, accuracy, and efficiency of patient management – with the potential to ultimately reduce diagnostic errors and improving clinical outcomes,” said Dr. Michael Lev, director of emergency radiology at Massachusetts General Hospital and professor of radiology at Harvard Medical School.

“MedyMatch is ideally positioned to leverage this technology, and their willingness to collaborate with industry partners reflects their awareness of, and sensitivity to, the complexities of patient assessment in the acute care setting,” Lev said. “The company’s first algorithms – CT detection of intracranial bleeds – represents the confluence of physician know-how and artificial intelligence clinical support.”

MedyMatch is currently conducting a clinical trial for its intracranial bleed assessment application and is working at getting FDA approval.

“MedyMatch can save patients millions of dollars in healthcare costs and prevent long-term suffering and treatment through early diagnosis,” said Tel Aviv based Zirra.com Ltd., in a report on the company. But because the company’s product hasn’t fully launched yet, is it still difficult “to assess the extent of market fit and user adoption.” Zirra is an Israel- based a research firm that analyses private companies using artificial intelligence and machine learning technologies.

MedyMatch has raised $2.2 million to date from investors including Exigent Ventures and Genesis Capital, according to Zirra, which values the company at $11 million to $15 million, based on a series of parameters including growth prospects. The company has an exit probability of 10 percent to 20%, the report said.