RECONSTRUCTION OF LOW-DOSE CTs USING ACCELERATED GENERATIVE MODELS BASED ON DIFFUSION TECHNIQUES

Currently, computed tomography (CT) is one of the modalities of choice for multiple pathologies due to its excellent diagnostic performance and speed of acquisition. However, it is also the radiological modality that is responsible for the highest irradiation from diagnostic studies, reaching radiation levels equivalent to a range between 6 months and 5.1 years of exposure to environmental radiation. Although radiation dose reduction strategies are implemented in most of the latest generation equipment, it is not always possible to effectively reduce patient irradiation, as it implies a decrease in image quality, which has negative consequences for diagnosis. On the other hand, the use of deep learning (DL) has spread in recent years in radiology. Thus, there are commercial products that apply DL to image noise reduction with promising results but which still fall into the limitation of a decrease in clinical performance when faced with aggressive radiation reductions.

In this context, we see that there is an opportunity in the development of a solution that, using state-of-the-art techniques in generative DL, allows the reconstruction of images obtained with a radiation dose reduction greater than 50% without compromising diagnostic performance. Given the recent scientific work on diffusion models applied to low-dose CT reconstruction, it is demonstrated that we start from already formulated technological concepts, which corresponds to a TRL 2 maturity level. Therefore, our goal at the end of the project is to reach a TRL 4 maturity level. That is, at least, we will reach a technology validated in laboratories.