Researchers developed a non-invasive technique to characterise lung cancer

Lung cancer is one of the most common and deadliest in the world. The biopsy is the method used for its diagnosis and characterisation, and although it allows obtaining valuable information, it could lead to certain clinical issues. The LuCaS (Lung cancer screening – A non-invasive methodology for early diagnosis) project aims to make decision support systems in the characterisation of lung cancer more objective and quantitative.

“The standard technique for assessing the mutation state of cancer resorts to biopsies, which generate highly reliable results, but are very invasive – in some cases leading to complications to the patients. Moreover, this technique is not able to characterise cancer globally, as only a portion of tissue is removed. An alternative solution is to make this characterisation using the radiological examinations through Computed Tomography, the same used for the initial diagnosis. Medical imaging allows obtaining a large set of useful information, according to a holistic approach comprehending a complete characterisation, providing opportunities to explore the relationship between the visual manifestations present in a medical image and the genetic profile of cancer, using a non-invasive approach”, explained Hélder Oliveira, a researcher at the Institute for Systems and Computer Engineering, Technology and Science (INESC TEC). “In this sense, the scope of the technology we are using will be more comprehensive than the biopsy itself, as it is based on three-dimensional data and is non-invasive. This way, it will also reduce costs significantly”, concluded the researcher who leads the project.

The project follows a “Radiogenomics” approach, i.e., it analyses the attributes of the image to describe and create mathematical models capable of identifying patterns and providing a prediction of the diagnosis, while relating characteristics of images with the analysis of genes collected during the biopsy. The project began in 2018, with several machine learning techniques developed thus far, using image information to predict the mutation state of lung cancer.

The project also comprised a prospective component, in which a model was developed to evaluate contributions of liquid biopsies in the characterisation of lung cancer. “This approach will be of great value, as a means to obtain molecular data in a minimally invasive way, compatible with the clinical processes,” said Hélder Oliveira.

The LuCaS (Lung cancer screening – A non-invasive methodology for early diagnosis) is currently under development, bringing together researchers from INESC TEC, the Faculty of Medicine of the University of Porto (FMUP), the Centro Hospitalar Universitário de São João and the Institute of Molecular Pathology and Immunology of the University of Porto – IPATIMUP at i3S. The project is co-funded by the COMPETE 2020 programme under the Scientific and Technological Research Support System – SAICT, with an eligible investment of €239K leading to an ERDF incentive of €203K.

The INESC TEC researcher mentioned in this news piece is associated with INESC TEC.