INESC TEC and the São João University Hospital (CHUSJ) developed a system for the diagnosis of pulmonary nodules. The innovation has been already tested in the radiology department of that hospital and allows the detection, characterisation and determination of the malignancy of pulmonary nodules.
This is the LNDetector project (System for the Automatic Detection, Segmentation and Classification of Pulmonary Nodules in Computed Tomography Images), which already gave rise to a prototype tested at CHUSJ and to an annotated database of Computed Tomography images, which contains a wide variety of nodules with different sizes, positions, shapes and textures, and the observation map of radiologists during the annotation process of exams.
It should be noted that there are already some methods to support the diagnosis in terms of detection, segmentation and classification of pulmonary nodules, but the technology, now developed by INESC TEC’s Centre for Biomedical Engineering Research (C-BER), also suggests the follow-up that the patient should follow.
The work of INESC TEC and CHUSJ therefore aims at improving the tools to support the diagnosis of lung cancer, which is the world’s deadliest cancer, but whose screening is a complex process due to the amount of images required and the possibility of failures in the identification and observation of pulmonary nodules.
The prototype developed – LNDetector (System for the Automatic Detection, Segmentation and Classification of Pulmonary Nodules in Computed Tomography Images) – received a funding of EUR 168,461 from the Foundation for Science and Technology (FCT), and has already won an award for best demo at the RECPAD 2019 conference, dedicated to the study of the areas of pattern recognition, image analysis and processing, computing and related areas.