EU initiative develops advanced diagnostics with light and AI

Led by the UPC's Centre for Sensors, Instruments and Systems Development, an international consortium of universities, hospitals and companies is developing new technologies based on light and AI to improve the diagnosis of several diseases.

Developing new photonics and AI tools for the early detection and appropriate treatment of ocular, cardiovascular and neurodegenerative diseases. This is the goal of the European project BE-LIGHT, coordinated by the Centre for Sensors, Instruments and Systems Development (CD6) of the Universitat Politècnica de Catalunya - BarcelonaTech (UPC) in collaboration with seven academic institutions, three hospitals and seven companies from Germany, France, Poland, Switzerland and Spain.

The consortium is led by Meritxell Vilaseca, a researcher at the CD6 and a professor at the Terrassa School of Optics and Optometry (FOOT) and the Barcelona School of Telecommunications Engineering (ETSETB), and Cristina Masoller, a researcher at the Nonlinear Dynamics, Nonlinear Optics and Lasers (DONLL) research group and a professor at the Terrassa School of Industrial, Aerospace and Audiovisual Engineering (ESEIAAT).

The project has received 2.5 million euros in funding from the EU's Horizon Europe program as part of the Marie Skłodowska-Curie Actions (Doctoral Networks).

Improving the diagnosis of retinal diseases

The BE-LIGHT project is a new step towards understanding and diagnosing retinal diseases, as explained by project coordinator Meritxell Vilaseca: "Through the use of neural networks and optogenetic techniques, the new tools will better understand the functioning of the retina and the exchange of information between neurons, and thus, based on quantitative models, detect diseases in the retina."

The researcher points out that "the combination of various photonic technologies, such as multispectral imaging and optical coherence tomography, complemented by AI algorithms, will allow for the precise analysis of ocular structures (cornea, vitreous, fundus) to detect visual diseases and oculomotor disorders early." In addition, according to Vilaseca, "the AI evaluation of the patterns of eye movements, strongly controlled by different brain regions, can also offer new diagnostic tools for neurological diseases, such as Alzheimer's or persistent Covid."

Cristina Masoller adds that "the project will also enable the development of new instruments and clinical methods capable of, for example, obtaining images of blood vessels or ocular structures of the patient through optical and optoacoustic tomography or thermal imaging." As the researcher explains, "this technology will help to non-invasively detect arteriosclerosis plaques early, as an example." Also, "new machine learning tools will be developed for light-based arrhythmia treatment and control, which could replace current techniques based on electrical impulses." Masoller adds that "the application of AI in combination with super-resolution microscopy techniques will make it possible to obtain images of biological structures less than one nanometer, such as proteins involved in Parkinson's disease and other rare diseases, improving their diagnosis".

Additionally, as part of the BE-LIGHT project, 11 researchers will be offered a training program in photonics, AI, machine learning, computer imaging and modeling.