A Canadian company is working on developing an artificial intelligence to detect signs of cancer in patients. According to the researchers, many lives will be saved if successful.
The issue of late diagnosis in cancer patients has been a subject of discussions for years. Millions of people around the world die every year because their cancer was diagnosed too late. The company Imagia created in 2014 in Montreal (Quebec) is seriously thinking about the concept of using artificial intelligence to automate the analysis of images via algorithms to detect the early signs of cancer.
“We expect to accelerate this process by letting the machine learn to analyze the images to associate them with specific genetic profiles, and possibly to choose the best treatment for the patient,” explained Alexandre Le Bouthillier, co-founder of Imagia in the columns of the local newspaper La Presse.
Imagia includes experts such as Lisa Di Jorio, a specialist in machine learning. Her role includes communicating with physicians and RN experts to develop tools in the health field.
“I need to be able to understand how to explain a clinical problem to an artificial intelligence researcher and, above all, I need to be able to explain to a clinician what is not yet possible,” explains the specialist.
Currently, artificial intelligences are not trained on medical data, but this should happen sometime in the future. Another challenge is that the data is not standardized across all hospitals — or even physicians. So, making sense of this data will not be obvious for an AI.
The research is still in its infancy, but despite the difficulties encountered today, researchers hope to save lives in a more or less near future. The goal is to be able to detect cancers faster, which will increase the chances of survival of patients.