Virtual Library
Start Your Search
Simon Padley
Author of
-
+
MA10 - Emerging Technologies for Lung Cancer Detection (ID 129)
- Event: WCLC 2019
- Type: Mini Oral Session
- Track: Screening and Early Detection
- Presentations: 1
- Now Available
- Moderators:Fabrizio Bianchi, Marcelo Sanchez
- Coordinates: 9/08/2019, 15:15 - 16:45, Dublin (1997)
-
+
MA10.10 - Uptake in Lung Cancer Screening – Does CT Location Matter? A Pilot Study Comparison of a Mobile and Hospital Based CT Scanner (Now Available) (ID 2165)
15:15 - 16:45 | Author(s): Simon Padley
- Abstract
- Presentation
Background
Community based lung cancer screening has been proposed as a method of increasing uptake for lung cancer screening by reducing barriers to participation. We report baseline statistics for a lung cancer screening pilot study in which patients were scanned on either a community based mobile CT unit or on a University Hospital based fixed-site CT scanner.
Method
Ever smokers aged 60-75 registered at 17 participating general practitioner practices (GP) in West London were invited for a lung health check at either a mobile unit situated in a supermarket car park or in a hospital site. The location offered was based upon proximity to the participant’s home address. On attendance a lung health check, assessing lung cancer risk, was undertaken. Participants with a LLPv2 score of ≥2.0% and/or PLCOM2012 score of ≥1.51% were offered a same day low dose CT (LDCT) scan. Uptake, attendance and non-attendance (DNA) rates were compared using Chi-squared (χ2) test.
Result
8366 potentially eligible participants were invited for a lung health check appointment; 5135 (61.4%) to the hospital site, and 3231 (38.6%) to the mobile site. 1749/8366 (20.9%) participants responded (males n=954/1749 (54.5%)). 1047/5135 (20.4%) were booked an appointment at the hospital site and 702/3231 (21.7%) at the mobile site (p=0.14). No difference was observed in lung cancer risk between participants at the two sites. Patients at the mobile site were more likely to be ex-smokers (p=0.048). The DNA rate at the hospital site was 96/1047 (9.2%) and at the mobile site was 48/702 (6.8%) (p=0.08). On attendance, 63 patients were ineligible for screening; 52/1749 (3.0%) did not meet the entry criteria and 11/1749 (0.6%) were acutely unwell. Therefore 1542 patients attended and had a risk score calculated and of these 1145/1542 (74.3%) underwent CT. Median [range] risk scores for scanned patients were 1.97 [0-25.34] for PLCOM2012 and 4.71 [0.94-35.92] for LLPv2. Lung cancer was confirmed in 17/1145 (1.5%) participants at baseline. A further 151/1145 (13.2%) participants will undergo interval CT for indeterminate nodules.
Conclusion
There was a small but non-significant increase in participant response rates for the community based mobile site compared to the hospital site CT scanner, but no difference in DNA rates. While community based mobile scanners may provide valuable additional capacity to lung screening programmes, the magnitude of any benefit to participant uptake needs to be balanced against the additional complexity of setting up these stand-alone facilities. Further work is ongoing to understand the interaction between CT location and other factors that influence recruitment, with a view to using effective methods to increase uptake at all sites for future screening invitations.
Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.
-
+
P2.11 - Screening and Early Detection (ID 178)
- Event: WCLC 2019
- Type: Poster Viewing in the Exhibit Hall
- Track: Screening and Early Detection
- Presentations: 1
- Now Available
- Moderators:
- Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall
-
+
P2.11-13 - What Is the Impact of Localised Data When Training Deep Neural Networks for Lung Cancer Prediction? (Now Available) (ID 2485)
10:15 - 18:15 | Author(s): Simon Padley
- Abstract
Background
Deep neural networks (DNN) have been shown to offer a viable alternative for risk cancer prediction of indeterminate pulmonary nodules (IPNs). While the type of data used for training is known to impact performance, this issue has not been extensively studied. We present, for the first time, a study of the effect of including training data that matches the clinical pathway of the independent validation dataset, a nodule clinic of incidental findings.
Method
Two identical DNNs were trained on the task of diagnosis prediction of pulmonary nodules from CT images. The first one (DNNnlst) used purely screening data from the US National Lung Screening Trial (922 cancer and 14733 benign nodules), while the second one (DNNnlst+incidental) included data of incidentally detected nodules from European hospitals (1064 cancer and 7207 benign nodules). Both models were evaluated in an independent validation set of nodules coming from a referral center in the UK (Royal Brompton and Harefield Hospital, London) consisting of baseline scans of 406 cancer and 325 benign nodules. The models were compared in terms of AUC, as well as their ability to reclassify cancer patients with intermediate risk nodules. The Intermediate risk sub-population was defined by selecting patients with nodules in the size range of 8 to 15mm, and who were followed-up within a year with CT, referred to PET-CT, or referred to biopsy. Within this sub-population, a cancer prevalence of 30% was assumed. The operating points of the cancer prediction models were chosen by setting a cancer risk of 70%, corresponding to high-risk nodules in the guidelines of the British Thoracic Society.
Result
The DNNnlst and DNNnlst+incidental models achieved an AUC of 84.33 (95%CI: 81.49, 87.15) and 87.43 (95%CI: 84.79, 89.82) respectively on the entire validation set, showing an improvement in the discrimination capabilities (p < 0.01). For reference, using the nodule’s maximum axial diameter as a predictor led to an AUC of 79.07 (95%CI: 75.73, 82.73). Additionally, considering only the intermediate risk population of the data, all of which would require workup according to guidelines, the DNNnlst+incidental model correctly classified as high risk 34.34% more cases than the DNNnlst model (sensitivity 59.39% (95%CI: 47.68, 75.17) vs. 44.21% (95%CI: 20.98, 64.42)), an improvement significant at p < 0.05.
Conclusion
Although a DNN trained only on the US lung cancer screening data could have clinical utility in an incidental setting, exposing it to further incidental data can not only increase its discriminability, as expected, but also make it a potentially more effective tool for speeding up the diagnosis of cancer patients with intermediate risk nodules and reducing unnecessary workups.
