Lung Cancer
Conditions
Keywords
Lung, Tumor, Artificial Intelligence
Brief summary
Observational cohort study involving individuals of both sexes with a history of smoking, residing in municipalities in the state of Bahia and attended by the mobile unit, with the aim of evaluating the integration of artificial intelligence (AI) in the detection of pulmonary nodules and the prediction of ASCT in high-risk individuals undergoing CT screening.
Detailed description
The main objective of the study is compare the performance of a Sybil AI tool with the LungRADS classifications assigned by radiologists for the risk stratification of pulmonary nodules. Furthermore, it aims to assess the correlation between the AI-predicted STAS and histopathological confirmation, alongside imaging and AI results. The hypothesis is that the Sybil AI model will demonstrate comparable predictive accuracy and that the features predicted by the AI will correlate with the presence of STAS.
Interventions
None listed
Sponsors
AstraZeneca
Study design
Observational model
COHORT
Time perspective
OTHER
Eligibility
Sex/Gender
ALL
Age
50 Years to 80 Years
Healthy volunteers
No
Inclusion criteria
* Individuals of both sexes, smokers or former smokers for a maximum of 15 years; * Smoking history of 20 pack-years or more; * Between 50 and 80 years old; * Provision of Free and Informed Consent in writing, signed and dated.
Exclusion criteria
* Individuals who are unable to undergo a CT scan; * Individuals who cannot tolerate lying on their back for more than 10 minutes continuously; * Individuals who present with symptoms highly suggestive of lung cancer (hemoptysis, chest pain, altered cough pattern, unintentional weight loss greater than 10 kg); * Diagnosis of severe heart disease while using multiple medications; * Diagnosis of severe lung disease, using multiple medications and/or requiring home oxygen therapy; * History of radiation therapy to the chest area; * Individuals undergoing cancer evaluation or treatment; * Individuals exhibiting signs of respiratory distress (nasal flaring, suprasternal retraction, use of accessory muscles, cyanosis); * Pregnancy.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Lung cancer risk stratification | through study completion, an average of 1 year | Measured by the agreement between the results of the Sybil AI model and the LungRADS classifications assigned by the radiologist |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Presence of STAs | through study completion, an average of 1 year | Conceptual definition: STAS is a histopathological finding in lung adenocarcinoma, in which tumor cells are observed disseminated in the alveolar spaces beyond the main tumor margin. o Operational definition: Presence of STAS confirmed by centralized histopathological review of lung tissue samples (biopsy or resection). The review is performed by pathologists who are unfamiliar with IA/radiological assessments |
| Histological diagnosis of lung cancer | through study completion, an average of 1 year | Confirmed by means of biopsy or surgical specimen with pathological subtyping. |
Countries
Brazil
Contacts
CONTACTAstraZeneca Clinical Study Information Center
PRINCIPAL_INVESTIGATORRicardo Sales dos Santos, Doctor
Fundação Bahiana de Cardiologia - FBC
Outcome results
None listed