Lung Cancer

Lung cancer is a phenomenon with global and public health issue implications. Several epidemiological studies highlight the relative risk factors and indicate new ways of diagnosis and treatments. This pilot within INCISIVE will address several significant challenges, such as enhancing sensitivity and specificity of low-cost imaging methods (as X-ray) to support earlier diagnosis, or enhancing lung cancer classification based on tumor cell morphology and histologic type characterization.

This pilot within INCISIVE will address several significant challenges, such as enhancing sensitivity and specificity of low-cost imaging methods (as X-ray) to support earlier diagnosis, or enhancing lung cancer classification based on tumor cell morphology and histologic type characterization.

Lung cancer has been transformed from a rare disease into a phenomenon with global and public health issue implications. Lung cancer is the second most common cancer in both men and women. More than 80% of individuals with lung cancer die either of the disease or due to the presentation of locally advanced or metastatic disease.

The pathogenetic mechanisms that may underlie lung cancer neoplasms are not well established, with industrialization, urbanization and environmental pollution playing crucial role. Several epidemiological studies of lung cancer highlight the relative risk factors and indicate new ways of diagnosis and treatments.

Screening programs are put in place to detect tumors in earlier, curable stages consequently reducing disease- specific mortality. A chest X-ray is usually the 1st test used to diagnose lung cancer. However, chest X-ray cannot give a definitive diagnosis using the CT scan as the next step. Additionally, the PET-CT scan may be done in order to better determine the early-stage cancer.

The challenges addressed in INCISIVE via this pilot are the following:

1. Recruitment challenged for screening approaches: Lung cancer screening is well established based on the most recent European protocol for the management of CT-detected nodules in Lancet Oncology in 2017. However, quality assurance in CT screening has been poorly implemented, offering room for improvement in several areas. Automated CT scan software and postprocessing procedures dedicated for lung cancer detection are attractive options.


2. Low-cost screening procedure is broadly applied in daily practical routine. Enhancing sensitivity and specificity of low-cost imaging methods (as X-ray) is crucial for achieving earlier diagnosis. Applying Artificial Intelligence algorithms on large datasets of CT images with known clinical outcomes will be used as our cornerstone for the development and validation of Lung Cancer imaging management tools.


3. Lung cancer classification based on tumor cell morphology and histologic type characterization will play a pivotal role in the diagnosis and management of lung cancer. AI tools for lung cancer diagnosis and classification based on histopathological outcome, in addition to low-cost screening procedures, further improve the therapeutic decision-making and effective prognostic outcome prediction in the era of personalized medicine.