A new technology has been developed in Russia that significantly reduces the radiation exposure to the patient during computed tomography (CT) scans. The innovation, presented by the AI company Smart Engines, is based on the use of artificial intelligence (AI) to control the scanning process.
The technology includes a neural network that monitors image quality in real time during the scan. Once the necessary amount of data for diagnosis has been collected, the AI stops the radiation, which reduces the radiation dose without reducing the accuracy of the diagnosis.
Traditionally, CT scans are performed according to a fixed protocol, regardless of the individual characteristics of the patient, which can lead to excessive radiation exposure. For example, a standard CT scan of the lungs requires about 10,000 X-rays. The new technology uses incomplete data to reconstruct the image, and AI analyzes it at each stage, completing the procedure as soon as enough information is available for diagnosis.
The method was tested primarily on the diagnosis of pneumonia, and the results showed that it reduces the radiation load by 15-25% without affecting the effectiveness of the diagnosis. Unlike standard protocols, the new method adapts to the characteristics of each patient, which not only reduces the radiation dose, but also minimizes the need for repeated scans, increasing the safety of the procedure.
