On an x-ray, even for an expert eye, it is difficult to detect the slightest lesion. In a clinic in Saint-Omer, in Hauts-de-France, the doctor spotted a fracture of the phalanx. Artificial intelligence, integrated into the software, allows it to confirm the diagnosis:
“If we ask the AI for its opinion, we click on it. She tells us yes, there is a fracture. And she frames this divide on the radio“, explains Dr Pierre Rialland, head of the clinic’s trauma department.
REPORT ARTICLE IA CLINIQUE SAINT OMER
AI to ensure correct diagnoses
Three types of lesions can thus be analyzed by AI. With 40 to 50 x-rays per day to process, for the medical team, it is a valuable tool that helps limit the risk of missing a fracture: “When in doubt, AI will confirm our diagnosis. Or not. It tells us whether there is a lesion, as we think, or not. And conversely, if we don’t really know if there is a fracture and the AI tells us that there is nothing, that reassures us. In the years to come, she could frankly be better than humans“, specifies Dr Pierre Rialland.
Reduction in emergency room waiting time
24,000 patients per year are admitted to trauma in this clinic. And the number is only increasing. For management, it is currently difficult to do without this technology, if only to reduce waiting times in emergency rooms.
More value-added procedures for radiologists
“Today we are in the shift of robotics, in the shift of artificial intelligence, said the head of the trauma department. And not doing so risks missing out on major developments. Especially since in fact, these technologies also have the advantage of saving time. And therefore, to allow the practitioner to refocus on their core business and work more on actions with higher added value.”
A first for this Saint-Omer clinic. The only one, for the moment, to test this AI-assisted software in the Hauts-de-France region. In total, fewer than 100 establishments in France, public and private, use it.
