Anagrafe della ricerca

A well-known biomarker of both oncological and autoimmune diseases is their vascular “fingerprint”. The regulation of vascular function in response to changing metabolic needs is essential for the maintenance of normal tissue and organ functions. In fact, one hallmark of cancer is the deregulation of angiogenic signaling, resulting in a process called neoangiogenesis, where the newly developed blood vessels are often organized chaotically and densely-packed around the tumor. Similarly, vascular dysregulation is also involved when considering autoimmune diseases and, in particular, inflammation, that is an important pathophysiological event that involves changes in cell composition and molecular characteristics, by promoting the infiltration of inflammatory immune cells and local oxidative stress and by triggering vascular pathological signals. Photoacoustics (PA) is a rapidly growing imaging modality that allows a non-invasive quantitative and metabolic analysis of vasculature. Studies have shown great potentials for this imaging modality, which is however still limited to ad-hoc specialized geometrical detectors or artefact-ridden images when using traditional linear probes for detection. The main goal of this project concerns with quantitative imaging of vasculature, and in particular vascular dysregulation, as a functional basis for autoimmune diseases and tumors using photoacoustic imaging and artificial intelligence. Innovative methods for PA volume imaging using a linear ultrasound probe will be developed and innovative multi-angle illumination of the imaged area will be explored for high-resolution and artefact-reduced volumetric and functional PA imaging. The project begins at in-silico and in-vitro levels, including simulations and phantom studies. Then, the activity will advance towards in-vivo chicken embryo studies. The project climaxes at the in-vivo clinical study of cancerous dermatological lesions and inflammation of the Achilles tendon. Importantly, artificial intelligence-based methods, specifically generative adversarial networks (GANs), will be employed to optimize and enhance the acquired photoacoustic images. Artificial intelligence methods will be employed to automatically segment and classify the acquired photoacoustic volumes and also the dermoscopic and histopathological biopsy images acquired in routine clinical practice. As a result of this project, an innovative 3D photoacoustic imaging system using a common ultrasound linear probe for detection will be developed, making the entire system potentially miniaturizable in the future for point-of-care solutions. The partnership of AI-VASCUES is a multidisciplinary team of experts in biomedical engineering, signal and image processing, together with clinicians in two fields. This composition will provide an ideal synergy of background knowledge, tech know-how, case scenarios and outreach potential for the best potential scientific and socioeconomic impact.