Israeli start-up Ibex helps detect cancer using AI
Ibex offers an important contribution in facing two challenges, the shortage of pathologists and the occurrence of human errors
As featured on The Jerusalem post
Tackling cancer is one of the most significant public health challenges of the 21st century, and accurately detecting the presence of tumors is an essential part of this challenge.Tel Aviv-based Ibex Medical Analytics supports medical professionals through its AI-based technology, which is designed to be used in routine clinical practice.
Ibex’s Galen Prostate solution helps identify suspected cancer on prostate core-needle biopsies. It just received the CE-IVD Mark, the company announced last Thursday, marking the first time one of its products has received the certification of conformity to EU health, safety and protection standards.
Ibex provides an important contribution in facing two challenges, the shortage of pathologists and the occurrence of human errors, CEO Joseph Mossel told The Jerusalem Post.
“When a person shows potential symptoms of cancer, the next step is a biopsy, an analysis of a sample of the affected tissues,” he said. “The examination of the sample has to be carried out by a pathologist using a microscope. Pathology is a complex medical discipline that requires years of training, and the world is facing a shortage of these specialists, even in countries such as the US and the UK. There is a growing gap between supply and demand.
”Galen Prostate offers a solution to support pathologists based on an algorithm that, among other benefits, provides them with a safety net to avoid a mistake that is not infrequent: tumors diagnosed as benign when they are in fact cancerous. In studies Ibex has carried out in collaboration with major labs in Israel, France and the US, such a mistake was registered at a rate between 3% and 12% of the cases, Mossel said.
“Labs perform their normal tests, and then they feed the results into our algorithm,” he said. “If it detects something unusual, an alert is issued.”
According to Ibex, the algorithm was based on a dataset of more than 60,000 prostate slides from multiple institutions with a variety of diagnoses and clinical findings.
“We achieved results that I would call unprecedented,” Mossel told the Post. “Blinded clinical trials that we carried out in Israel and in France, testing the findings of our technology compared to those of several pathologists examining each sample, showed that our algorithm was able to detect cancer in over 99% of the cases when it was present.”
“It is important for us to highlight that we have already prevented mistakes,” he said. “We have already helped people get a correct diagnosis. Discovering to be sick with cancer is not good news, but the alternative of a mistake in the assessment of the situation is much worse.”
Dr. Delphine Raoux, the head of innovation technologies at Medipath, the largest network of private pathology labs in France, said: “We are impressed with the results of the study, particularly with the performance of Galen Prostate, which was in near-perfect concordance with the consensus diagnosis between multiple pathologists. This solution can become an important tool for pathologists that look for ways to reduce diagnostic error rates.”
“Medipath believes that AI-powered solutions are the future of pathology, and we are happy to participate in trials that demonstrate their accuracy and utility and to collaborate with Ibex to help the evolution of their AI solutions,” she said, according to an Ibex press release.
Galen Prostate is ready to be commercialized, and several clients are waiting for the CE-IVD Mark, Mossel said.
“We are the only company in our field ready to sell our product,” he said.
Ibex has developed an algorithm for detecting breast cancer, and it already has been used in labs, Mossel said.
“We are hoping it will be CE-approved by the end of the year,” he said.Ibex believes its technology is potentially relevant for all types of cancer.
“We are also working on a product whose focus is helping pathologists be more efficient and accelerate the examination, which will help address the problem of the shortage of specialists” Mossel said.