McCabe, J;
Cheng, D;
Bhamani, A;
Mullin, M;
Patrick, T;
Nair, A;
Janes, SM;
... Jacob, J; + view all
(2024)
Exploring Fairness in State-of-the-Art Pulmonary Nodule Detection Algorithms.
In: Puyol-Antón, Esther and Zamzmi, Ghada and Feragen, Aasa and King, Andrew P and Cheplygina, Veronika and Ganz-Benjaminsen, Melanie and Ferrante, Enzo and Glocker, Ben and Petersen, Eike and Baxter, John SH and Rekik, Islem and Eagleson, Roy, (eds.)
Ethics and Fairness in Medical Imaging: FAIMI EPIMI 2024.
(pp. pp. 78-87).
Springer, Cham
![[thumbnail of Fairness_In_SOTA_Nodule_Detection_Algorithms.pdf]](https://discovery.ucl.ac.uk/style/images/fileicons/text.png) |
Text
Fairness_In_SOTA_Nodule_Detection_Algorithms.pdf - Accepted Version Access restricted to UCL open access staff until 14 October 2025. Download (720kB) |
Abstract
Lung cancer is the leading cause of cancer mortality worldwide. Asymptomatic in its early stages, it is disproportionately detected when the disease is advanced. Resource constraints have resulted in increasing reliance on computer-aided detection (CADe) systems to assist with scan evaluation. The datasets used to train these algorithms are often unbalanced in their representation of protected groups e.g. sex and ethnicity. This project investigates whether there are performance disparities in detecting clinically relevant nodules across under-represented groups in selected, state-of-the-art nodule detection algorithms trained on data from a screening program in the UK. Our analysis revealed that overall, the algorithms demonstrate equitable performance across various demographic groups. However, their performance varies strongly across nodule characteristics (size and type) in line with their prevalence in the training set. To ensure continued equitable performance, algorithms should not only consider demographic but also nodule attributes representativeness in their training.
Loading...
Loading...Loading...Loading...Archive Staff Only
 |
View Item |
