Abstract 104P
Background
The COVID-19 pandemic has made the patient journey very difficult especially for diagnostics of new lung cancer (LC) cases because of lockdown, social distancing, similarity of symptoms and limitations with healthcare access. Аt the same time thousands of patients underwent CT for COVID detection. The aim of our study was to assess AI technology for LC detection in a COVID CT scan database.
Methods
Chest CT scans (without age, sex, smoking history, COVID-19 severity grade and other limitations) were retrospectively anonymized and analyzed by AI platform (BotkinAI). All findings were classified according to Lung-RADS criteria, reassessed by experts, patients were checked with regional onco registry and, if necessary, followed up for LC confirmation.
Results
CT scans from 9709 patients were analyzed: 9524 (98%) cases were Lung-RADS 0-2 and 185 (2%) were Lung-RADS 4+. Among Lung-RADS 4+ cases 58 (31%) patients had previously diagnosed LC or other malignancies with lung metastases, 5 (2.7%) pts were newly diagnosed (4 had LC and 1 metastatic breast cancer), 23 (12.4%) cases were false positive (LC was not confirmed), 74 (40%) patients refused from follow up diagnostics procedure or were unavailable, 25 (13.5%) pts has died and. Among 9524 Lung-RADS 0-2 cases 3 (0.03%) patients were false negative and had confirmed lung cancer. The sensitivity of AI was 95.4% (95% CI 87.29%-99.05%), specificity- 99.7% (95% CI 99.64% - 99.85%), concordance- 99.7% (95% CI 99.60% - 99.82%), positive predictive value - 73.2% (95% CI 64.48% - 80.52%), negative predictive value - 99.9% (95% CI 99.90% - 99.99%) respectively.
Conclusions
The COVID-19 pandemic gave a unique chance for incidental pulmonary nodes detection and artificial intelligent technology can provide support to physicians, save time and increase effectiveness of CT scans analysis.
Legal entity responsible for the study
The author.
Funding
AstraZeneca.
Disclosure
The author has declared no conflicts of interest.