Researchers at the Institute of Computer Systems and Engineering, Technology and Science (INESC TEC), in Porto, have developed a non-invasive technique that, using three-dimensional analysis of medical images, enables the characterization of lung cancer, was announced today.
In a press release, the Porto institute specifies that the non-invasive technique was developed within the framework of the LUCAS project, whose objective is to make decision support systems for the characterization of lung cancer “more objective and quantitative.
In addition to researchers from INESC TEC, the project includes specialists from the Faculty of Medicine of Porto (FMUP), the Centro Hospitalar Universitário de São João (CHUSJ) and the Institute of Molecular Pathology and Immunology of the University of Porto (Ipatimup).
The method developed uses three-dimensional analysis of the medical image – computed tomography – to “describe and create mathematical models capable of identifying patterns and offering a prediction of diagnosis”, linking the characteristics of the image to the analysis of genes collected during biopsy, a plus used for diagnosis and characterization.
Quoted in the press release, Helder Oliveira, researcher at INESC TEC and project leader, specifies that the “standard” technique for assessing the status of cancer uses biopsy, a method which “produces very reliable results, but which is quite invasive and which, in some cases, can lead to complications”. to the patient”.
The biopsy, explains the researcher, “does not make it possible to characterize the cancer globally, because only a part of the tissues is removed”.
Helder Oliveira explains that medical imaging makes it possible to obtain “a wide range of useful information”, opening up opportunities to study the relationship between the visual manifestations present in the medical image and the genetic profile of cancer.
“The technology we use will have a broader reach than the biopsy itself, as it is based on three-dimensional information and is non-invasive,” he observes, noting that the method “dramatically reduces costs.” .
As part of the project, which began in 2018, researchers have already developed “machine learning” techniques that use image information to predict the mutated state of lung cancer.
In parallel with the development of the new technique, the project integrated a “prospective component”, in which a model was developed to assess the contributions of liquid biopsy in lung characterization.
“This approach will be of great value as a way to obtain molecular data in a minimally invasive way that is compatible with clinical routine,” adds Helder Oliveira.
The project is co-financed by the Compete 2020 program within the framework of the Support System for Scientific and Technological Research (SAICT), in an eligible investment of 239 thousand euros.