Israeli AI Company Develops Tool to Predict Alzheimer's Risk Years Before Symptoms Appear


Developed from research conducted at Beilinson Hospital in collaboration with Ariel University, the AI model developed by Israeli company ALZAI Health has reached a major milestone, with validation involving more than 1.6 million medical records from the United States and Israel. The technology could help identify individuals at high risk of developing Alzheimer's disease long before the first signs of cognitive decline appear.
The model analyzes routine blood test results, medical histories, existing diagnoses, and selected demographic information to assess an individual's risk of developing the disease. Its main advantage is that it does not require expensive brain scans or invasive procedures at the initial screening stage. Instead, it uses information already available in patients' electronic medical records.
Detecting Alzheimer's Before the First Symptoms
One of the main challenges in combating Alzheimer's disease is identifying the condition early enough to allow for appropriate medical intervention.
The disease develops gradually. The first biological changes in the brain can occur 15 to 20 years before clinical symptoms emerge. During this period, amyloid plaques and tau protein aggregates progressively accumulate in the brain. By the time memory problems and cognitive difficulties become noticeable, significant damage may already have occurred.
Early detection is becoming increasingly important with the arrival of new treatments whose effectiveness has primarily been demonstrated during the early stages of the disease. Yet many patients are not diagnosed until cognitive decline is already established.
"Early diagnosis is essential because it offers an opportunity to slow the progression of Alzheimer's disease. Some of the treatments available today are only effective in the early stages. Throughout my years of medical practice, I have repeatedly encountered patients who only sought medical attention after developing significant cognitive difficulties. At that point, medicine has far fewer options available," explains Dr. Amir Glik, head of the Cognitive Neurology Department at Beilinson Hospital, who initiated the research.
From Medical Research to an International Technology Innovation
The project was initiated by Dr. Amir Glik, together with Professor Chen Hajjaj, Dr. Anat Goldstein, and Dr. Orit Refaeli, from the Department of Industrial Engineering and Management at Ariel University.
Drawing on Clalit's extensive medical databases and Beilinson Hospital's research capabilities, the team analyzed hundreds of thousands of patient records. Researchers examined routine blood test results, pre-existing medical conditions, and demographic information.
Using machine learning models trained on hundreds of millions of medical data points, they identified indicators associated with the future risk of developing Alzheimer's disease. Individually, these factors may not necessarily raise any particular medical concern. However, artificial intelligence can identify potentially significant patterns by analyzing combinations of these indicators.
This research laid the scientific foundations for the creation of ALZAI Health. The company was established through a partnership between Mor Research Applications, Clalit's technology transfer company; Ariel Scientific Innovation, its equivalent at Ariel University; and the company's founders: CEO Haim Roklow, along with Roi Kite and Ofri Kite.

Since then, the startup has expanded beyond its original research environment. Its international validation study analyzed more than 1.6 million medical records and over 200 million data points from Israel and the United States.
ALZAI Health also began trading on the Toronto Stock Exchange's TSX Venture Exchange approximately two months before the announcement and has raised nearly CAD 7.5 million to date.
The company's next ambition is to make its risk assessment tool available to healthcare systems, pharmaceutical companies, and clinical research programs worldwide.
A Global Public Health Challenge
According to the World Health Organization, nearly 60 million people worldwide are currently living with dementia. Alzheimer's disease accounts for approximately 60% to 70% of all cases. With populations aging, this figure could exceed 150 million by 2050.
In this context, the ability to identify individuals at increased risk using data already collected during routine medical consultations could pave the way for a more accessible approach, with the potential for large-scale implementation.
For Dr. Erez Barenboim, CEO of Clalit-Beilinson and Hasharon Hospitals, this development highlights the role of hospital-based research in shaping the future of medicine.
For Hayim Raclaw, co-founder, CEO, and director of ALZAI Health Group, the objective is to transform how healthcare systems approach the disease.
"Our vision is to move from a model of medicine that responds to disease only after symptoms appear to an approach capable of identifying at-risk individuals much earlier," he explains.
He adds that the technology's main advantage lies in its ability to use information already available in medical records, without initially requiring complex or invasive examinations.
While the technology still needs to find its place in clinical practice, it opens up an important possibility: earlier detection of neurodegenerative diseases, at a stage when every year gained could make a difference.
Caroline Haïat





Comments